diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2020-10-20 13:26:30 -0700 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2020-10-20 13:26:30 -0700 |
| commit | 59f0e7eb2f9ffa7715ca95908797b52ba35af11a (patch) | |
| tree | 9e6bd8cf33f3b92634efb6f1577e0c121cd2b696 | |
| parent | 4962a85696f9439970bfd84f7ce23b2721f13549 (diff) | |
| parent | 8c39076c276be0b31982e44654e2c2357473258a (diff) | |
| download | linux-stable-59f0e7eb2f9ffa7715ca95908797b52ba35af11a.tar.gz linux-stable-59f0e7eb2f9ffa7715ca95908797b52ba35af11a.tar.bz2 linux-stable-59f0e7eb2f9ffa7715ca95908797b52ba35af11a.zip | |
Merge tag 'nfs-for-5.10-1' of git://git.linux-nfs.org/projects/anna/linux-nfs
Pull NFS client updates from Anna Schumaker:
"Stable Fixes:
- Wait for stateid updates after CLOSE/OPEN_DOWNGRADE # v5.4+
- Fix nfs_path in case of a rename retry
- Support EXCHID4_FLAG_SUPP_FENCE_OPS v4.2 EXCHANGE_ID flag
New features and improvements:
- Replace dprintk() calls with tracepoints
- Make cache consistency bitmap dynamic
- Added support for the NFS v4.2 READ_PLUS operation
- Improvements to net namespace uniquifier
Other bugfixes and cleanups:
- Remove redundant clnt pointer
- Don't update timeout values on connection resets
- Remove redundant tracepoints
- Various cleanups to comments
- Fix oops when trying to use copy_file_range with v4.0 source server
- Improvements to flexfiles mirrors
- Add missing 'local_lock=posix' mount option"
* tag 'nfs-for-5.10-1' of git://git.linux-nfs.org/projects/anna/linux-nfs: (55 commits)
NFSv4.2: support EXCHGID4_FLAG_SUPP_FENCE_OPS 4.2 EXCHANGE_ID flag
NFSv4: Fix up RCU annotations for struct nfs_netns_client
NFS: Only reference user namespace from nfs4idmap struct instead of cred
nfs: add missing "posix" local_lock constant table definition
NFSv4: Use the net namespace uniquifier if it is set
NFSv4: Clean up initialisation of uniquified client id strings
NFS: Decode a full READ_PLUS reply
SUNRPC: Add an xdr_align_data() function
NFS: Add READ_PLUS hole segment decoding
SUNRPC: Add the ability to expand holes in data pages
SUNRPC: Split out _shift_data_right_tail()
SUNRPC: Split out xdr_realign_pages() from xdr_align_pages()
NFS: Add READ_PLUS data segment support
NFS: Use xdr_page_pos() in NFSv4 decode_getacl()
SUNRPC: Implement a xdr_page_pos() function
SUNRPC: Split out a function for setting current page
NFS: fix nfs_path in case of a rename retry
fs: nfs: return per memcg count for xattr shrinkers
NFSv4: Wait for stateid updates after CLOSE/OPEN_DOWNGRADE
nfs: remove incorrect fallthrough label
...
36 files changed, 989 insertions, 502 deletions
diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c index 1eabd91870e6..1d9488cf0534 100644 --- a/fs/lockd/mon.c +++ b/fs/lockd/mon.c @@ -417,7 +417,7 @@ void nsm_release(struct nsm_handle *nsm) /* * XDR functions for NSM. * - * See http://www.opengroup.org/ for details on the Network + * See https://www.opengroup.org/ for details on the Network * Status Monitor wire protocol. */ diff --git a/fs/nfs/fs_context.c b/fs/nfs/fs_context.c index 222afba70bc0..29ec8b09a52d 100644 --- a/fs/nfs/fs_context.c +++ b/fs/nfs/fs_context.c @@ -94,6 +94,7 @@ enum { static const struct constant_table nfs_param_enums_local_lock[] = { { "all", Opt_local_lock_all }, { "flock", Opt_local_lock_flock }, + { "posix", Opt_local_lock_posix }, { "none", Opt_local_lock_none }, {} }; diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index 6b063227e34e..2bcbe38afe2e 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -32,9 +32,9 @@ int nfs_mountpoint_expiry_timeout = 500 * HZ; /* * nfs_path - reconstruct the path given an arbitrary dentry * @base - used to return pointer to the end of devname part of path - * @dentry - pointer to dentry + * @dentry_in - pointer to dentry * @buffer - result buffer - * @buflen - length of buffer + * @buflen_in - length of buffer * @flags - options (see below) * * Helper function for constructing the server pathname @@ -49,15 +49,19 @@ int nfs_mountpoint_expiry_timeout = 500 * HZ; * the original device (export) name * (if unset, the original name is returned verbatim) */ -char *nfs_path(char **p, struct dentry *dentry, char *buffer, ssize_t buflen, - unsigned flags) +char *nfs_path(char **p, struct dentry *dentry_in, char *buffer, + ssize_t buflen_in, unsigned flags) { char *end; int namelen; unsigned seq; const char *base; + struct dentry *dentry; + ssize_t buflen; rename_retry: + buflen = buflen_in; + dentry = dentry_in; end = buffer+buflen; *--end = '\0'; buflen--; diff --git a/fs/nfs/nfs42xattr.c b/fs/nfs/nfs42xattr.c index 86777996cfec..b51424ff8159 100644 --- a/fs/nfs/nfs42xattr.c +++ b/fs/nfs/nfs42xattr.c @@ -67,7 +67,6 @@ struct nfs4_xattr_bucket { struct nfs4_xattr_cache { struct kref ref; - spinlock_t hash_lock; /* protects hashtable and lru */ struct nfs4_xattr_bucket buckets[NFS4_XATTR_HASH_SIZE]; struct list_head lru; struct list_head dispose; @@ -882,7 +881,7 @@ nfs4_xattr_cache_count(struct shrinker *shrink, struct shrink_control *sc) { unsigned long count; - count = list_lru_count(&nfs4_xattr_cache_lru); + count = list_lru_shrink_count(&nfs4_xattr_cache_lru, sc); return vfs_pressure_ratio(count); } @@ -976,7 +975,7 @@ nfs4_xattr_entry_count(struct shrinker *shrink, struct shrink_control *sc) lru = (shrink == &nfs4_xattr_large_entry_shrinker) ? &nfs4_xattr_large_entry_lru : &nfs4_xattr_entry_lru; - count = list_lru_count(lru); + count = list_lru_shrink_count(lru, sc); return vfs_pressure_ratio(count); } diff --git a/fs/nfs/nfs42xdr.c b/fs/nfs/nfs42xdr.c index cc50085e151c..0dc31ad2362e 100644 --- a/fs/nfs/nfs42xdr.c +++ b/fs/nfs/nfs42xdr.c @@ -45,6 +45,15 @@ #define encode_deallocate_maxsz (op_encode_hdr_maxsz + \ encode_fallocate_maxsz) #define decode_deallocate_maxsz (op_decode_hdr_maxsz) +#define encode_read_plus_maxsz (op_encode_hdr_maxsz + \ + encode_stateid_maxsz + 3) +#define NFS42_READ_PLUS_SEGMENT_SIZE (1 /* data_content4 */ + \ + 2 /* data_info4.di_offset */ + \ + 2 /* data_info4.di_length */) +#define decode_read_plus_maxsz (op_decode_hdr_maxsz + \ + 1 /* rpr_eof */ + \ + 1 /* rpr_contents count */ + \ + 2 * NFS42_READ_PLUS_SEGMENT_SIZE) #define encode_seek_maxsz (op_encode_hdr_maxsz + \ encode_stateid_maxsz + \ 2 /* offset */ + \ @@ -128,6 +137,14 @@ decode_putfh_maxsz + \ decode_deallocate_maxsz + \ decode_getattr_maxsz) +#define NFS4_enc_read_plus_sz (compound_encode_hdr_maxsz + \ + encode_sequence_maxsz + \ + encode_putfh_maxsz + \ + encode_read_plus_maxsz) +#define NFS4_dec_read_plus_sz (compound_decode_hdr_maxsz + \ + decode_sequence_maxsz + \ + decode_putfh_maxsz + \ + decode_read_plus_maxsz) #define NFS4_enc_seek_sz (compound_encode_hdr_maxsz + \ encode_sequence_maxsz + \ encode_putfh_maxsz + \ @@ -324,6 +341,16 @@ static void encode_deallocate(struct xdr_stream *xdr, encode_fallocate(xdr, args); } +static void encode_read_plus(struct xdr_stream *xdr, + const struct nfs_pgio_args *args, + struct compound_hdr *hdr) +{ + encode_op_hdr(xdr, OP_READ_PLUS, decode_read_plus_maxsz, hdr); + encode_nfs4_stateid(xdr, &args->stateid); + encode_uint64(xdr, args->offset); + encode_uint32(xdr, args->count); +} + static void encode_seek(struct xdr_stream *xdr, const struct nfs42_seek_args *args, struct compound_hdr *hdr) @@ -723,6 +750,28 @@ static void nfs4_xdr_enc_deallocate(struct rpc_rqst *req, } /* + * Encode READ_PLUS request + */ +static void nfs4_xdr_enc_read_plus(struct rpc_rqst *req, + struct xdr_stream *xdr, + const void *data) +{ + const struct nfs_pgio_args *args = data; + struct compound_hdr hdr = { + .minorversion = nfs4_xdr_minorversion(&args->seq_args), + }; + + encode_compound_hdr(xdr, req, &hdr); + encode_sequence(xdr, &args->seq_args, &hdr); + encode_putfh(xdr, args->fh, &hdr); + encode_read_plus(xdr, args, &hdr); + + rpc_prepare_reply_pages(req, args->pages, args->pgbase, + args->count, hdr.replen); + encode_nops(&hdr); +} + +/* * Encode SEEK request */ static void nfs4_xdr_enc_seek(struct rpc_rqst *req, @@ -970,6 +1019,97 @@ static int decode_deallocate(struct xdr_stream *xdr, struct nfs42_falloc_res *re return decode_op_hdr(xdr, OP_DEALLOCATE); } +static int decode_read_plus_data(struct xdr_stream *xdr, struct nfs_pgio_res *res, + uint32_t *eof) +{ + uint32_t count, recvd; + uint64_t offset; + __be32 *p; + + p = xdr_inline_decode(xdr, 8 + 4); + if (unlikely(!p)) + return -EIO; + + p = xdr_decode_hyper(p, &offset); + count = be32_to_cpup(p); + recvd = xdr_align_data(xdr, res->count, count); + res->count += recvd; + + if (count > recvd) { + dprintk("NFS: server cheating in read reply: " + "count %u > recvd %u\n", count, recvd); + *eof = 0; + return 1; + } + + return 0; +} + +static int decode_read_plus_hole(struct xdr_stream *xdr, struct nfs_pgio_res *res, + uint32_t *eof) +{ + uint64_t offset, length, recvd; + __be32 *p; + + p = xdr_inline_decode(xdr, 8 + 8); + if (unlikely(!p)) + return -EIO; + + p = xdr_decode_hyper(p, &offset); + p = xdr_decode_hyper(p, &length); + recvd = xdr_expand_hole(xdr, res->count, length); + res->count += recvd; + + if (recvd < length) { + *eof = 0; + return 1; + } + return 0; +} + +static int decode_read_plus(struct xdr_stream *xdr, struct nfs_pgio_res *res) +{ + uint32_t eof, segments, type; + int status, i; + __be32 *p; + + status = decode_op_hdr(xdr, OP_READ_PLUS); + if (status) + return status; + + p = xdr_inline_decode(xdr, 4 + 4); + if (unlikely(!p)) + return -EIO; + + eof = be32_to_cpup(p++); + segments = be32_to_cpup(p++); + if (segments == 0) + goto out; + + for (i = 0; i < segments; i++) { + p = xdr_inline_decode(xdr, 4); + if (unlikely(!p)) + return -EIO; + + type = be32_to_cpup(p++); + if (type == NFS4_CONTENT_DATA) + status = decode_read_plus_data(xdr, res, &eof); + else if (type == NFS4_CONTENT_HOLE) + status = decode_read_plus_hole(xdr, res, &eof); + else + return -EINVAL; + + if (status < 0) + return status; + if (status > 0) + break; + } + +out: + res->eof = eof; + return 0; +} + static int decode_seek(struct xdr_stream *xdr, struct nfs42_seek_res *res) { int status; @@ -1147,6 +1287,33 @@ out: } /* + * Decode READ_PLUS request + */ +static int nfs4_xdr_dec_read_plus(struct rpc_rqst *rqstp, + struct xdr_stream *xdr, + void *data) +{ + struct nfs_pgio_res *res = data; + struct compound_hdr hdr; + int status; + + status = decode_compound_hdr(xdr, &hdr); + if (status) + goto out; + status = decode_sequence(xdr, &res->seq_res, rqstp); + if (status) + goto out; + status = decode_putfh(xdr); + if (status) + goto out; + status = decode_read_plus(xdr, res); + if (!status) + status = res->count; +out: + return status; +} + +/* * Decode SEEK request */ static int nfs4_xdr_dec_seek(struct rpc_rqst *rqstp, diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 0c9505dc852c..065cb04222a1 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -599,6 +599,14 @@ static inline bool nfs4_stateid_is_newer(const nfs4_stateid *s1, const nfs4_stat return (s32)(be32_to_cpu(s1->seqid) - be32_to_cpu(s2->seqid)) > 0; } +static inline bool nfs4_stateid_is_next(const nfs4_stateid *s1, const nfs4_stateid *s2) +{ + u32 seq1 = be32_to_cpu(s1->seqid); + u32 seq2 = be32_to_cpu(s2->seqid); + + return seq2 == seq1 + 1U || (seq2 == 1U && seq1 == 0xffffffffU); +} + static inline bool nfs4_stateid_match_or_older(const nfs4_stateid *dst, const nfs4_stateid *src) { return nfs4_stateid_match_other(dst, src) && diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index daacc78a3d48..be7915c861ce 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -1045,6 +1045,8 @@ static int nfs4_server_common_setup(struct nfs_server *server, server->caps |= server->nfs_client->cl_mvops->init_caps; if (server->flags & NFS_MOUNT_NORDIRPLUS) server->caps &= ~NFS_CAP_READDIRPLUS; + if (server->nfs_client->cl_proto == XPRT_TRANSPORT_RDMA) + server->caps &= ~NFS_CAP_READ_PLUS; /* * Don't use NFS uid/gid mapping if we're using AUTH_SYS or lower * authentication. diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c index fdfc77486ace..91be7f628e4a 100644 --- a/fs/nfs/nfs4file.c +++ b/fs/nfs/nfs4file.c @@ -145,7 +145,8 @@ static ssize_t __nfs4_copy_file_range(struct file *file_in, loff_t pos_in, /* Only offload copy if superblock is the same */ if (file_in->f_op != &nfs4_file_operations) return -EXDEV; - if (!nfs_server_capable(file_inode(file_out), NFS_CAP_COPY)) + if (!nfs_server_capable(file_inode(file_out), NFS_CAP_COPY) || + !nfs_server_capable(file_inode(file_in), NFS_CAP_COPY)) return -EOPNOTSUPP; if (file_inode(file_in) == file_inode(file_out)) return -EOPNOTSUPP; diff --git a/fs/nfs/nfs4idmap.c b/fs/nfs/nfs4idmap.c index 62e6eea5c516..8d8aba305ecc 100644 --- a/fs/nfs/nfs4idmap.c +++ b/fs/nfs/nfs4idmap.c @@ -46,6 +46,7 @@ #include <keys/user-type.h> #include <keys/request_key_auth-type.h> #include <linux/module.h> +#include <linux/user_namespace.h> #include "internal.h" #include "netns.h" @@ -69,13 +70,13 @@ struct idmap { struct rpc_pipe *idmap_pipe; struct idmap_legacy_upcalldata *idmap_upcall_data; struct mutex idmap_mutex; - const struct cred *cred; + struct user_namespace *user_ns; }; static struct user_namespace *idmap_userns(const struct idmap *idmap) { - if (idmap && idmap->cred) - return idmap->cred->user_ns; + if (idmap && idmap->user_ns) + return idmap->user_ns; return &init_user_ns; } @@ -286,7 +287,7 @@ static struct key *nfs_idmap_request_key(const char *name, size_t namelen, if (ret < 0) return ERR_PTR(ret); - if (!idmap->cred || idmap->cred->user_ns == &init_user_ns) + if (!idmap->user_ns || idmap->user_ns == &init_user_ns) rkey = request_key(&key_type_id_resolver, desc, ""); if (IS_ERR(rkey)) { mutex_lock(&idmap->idmap_mutex); @@ -462,7 +463,7 @@ nfs_idmap_new(struct nfs_client *clp) return -ENOMEM; mutex_init(&idmap->idmap_mutex); - idmap->cred = get_cred(clp->cl_rpcclient->cl_cred); + idmap->user_ns = get_user_ns(clp->cl_rpcclient->cl_cred->user_ns); rpc_init_pipe_dir_object(&idmap->idmap_pdo, &nfs_idmap_pipe_dir_object_ops, @@ -486,7 +487,7 @@ nfs_idmap_new(struct nfs_client *clp) err_destroy_pipe: rpc_destroy_pipe_data(idmap->idmap_pipe); err: - put_cred(idmap->cred); + get_user_ns(idmap->user_ns); kfree(idmap); return error; } @@ -503,7 +504,7 @@ 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); + put_user_ns(idmap->user_ns); kfree(idmap); } diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 6e95c85fe395..9e0ca9b2b210 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -63,6 +63,7 @@ #include "callback.h" #include "pnfs.h" #include "netns.h" +#include "sysfs.h" #include "nfs4idmap.h" #include "nfs4session.h" #include "fscache.h" @@ -70,6 +71,10 @@ #include "nfs4trace.h" +#ifdef CONFIG_NFS_V4_2 +#include "nfs42.h" +#endif /* CONFIG_NFS_V4_2 */ + #define NFSDBG_FACILITY NFSDBG_PROC #define NFS4_BITMASK_SZ 3 @@ -107,6 +112,9 @@ static int nfs41_test_stateid(struct nfs_server *, nfs4_stateid *, static int nfs41_free_stateid(struct nfs_server *, const nfs4_stateid *, const struct cred *, bool); #endif +static void nfs4_bitmask_adjust(__u32 *bitmask, struct inode *inode, + struct nfs_server *server, + struct nfs4_label *label); #ifdef CONFIG_NFS_V4_SECURITY_LABEL static inline struct nfs4_label * @@ -1547,19 +1555,6 @@ static void nfs_state_log_update_open_stateid(struct nfs4_state *state) wake_up_all(&state->waitq); } -static void nfs_state_log_out_of_order_open_stateid(struct nfs4_state *state, - const nfs4_stateid *stateid) -{ - u32 state_seqid = be32_to_cpu(state->open_stateid.seqid); - u32 stateid_seqid = be32_to_cpu(stateid->seqid); - - if (stateid_seqid == state_seqid + 1U || - (stateid_seqid == 1U && state_seqid == 0xffffffffU)) - nfs_state_log_update_open_stateid(state); - else - set_bit(NFS_STATE_CHANGE_WAIT, &state->flags); -} - static void nfs_test_and_clear_all_open_stateid(struct nfs4_state *state) { struct nfs_client *clp = state->owner->so_server->nfs_client; @@ -1585,21 +1580,19 @@ static void nfs_test_and_clear_all_open_stateid(struct nfs4_state *state) * i.e. The stateid seqids have to be initialised to 1, and * are then incremented on every state transition. */ -static bool nfs_need_update_open_stateid(struct nfs4_state *state, +static bool nfs_stateid_is_sequential(struct nfs4_state *state, const nfs4_stateid *stateid) { - if (test_bit(NFS_OPEN_STATE, &state->flags) == 0 || - !nfs4_stateid_match_other(stateid, &state->open_stateid)) { + if (test_bit(NFS_OPEN_STATE, &state->flags)) { + /* The common case - we're updating to a new sequence number */ + if (nfs4_stateid_match_other(stateid, &state->open_stateid) && + nfs4_stateid_is_next(&state->open_stateid, stateid)) { + return true; + } + } else { + /* This is the first OPEN in this generation */ if (stateid->seqid == cpu_to_be32(1)) - nfs_state_log_update_open_stateid(state); - else - set_bit(NFS_STATE_CHANGE_WAIT, &state->flags); - return true; - } - - if (nfs4_stateid_is_newer(stateid, &state->open_stateid)) { - nfs_state_log_out_of_order_open_stateid(state, stateid); - return true; + return true; } return false; } @@ -1673,16 +1666,16 @@ static void nfs_set_open_stateid_locked(struct nfs4_state *state, int status = 0; for (;;) { - if (!nfs_need_update_open_stateid(state, stateid)) - return; - if (!test_bit(NFS_STATE_CHANGE_WAIT, &state->flags)) + if (nfs_stateid_is_sequential(state, stateid)) break; + if (status) break; /* Rely on seqids for serialisation with NFSv4.0 */ if (!nfs4_has_session(NFS_SERVER(state->inode)->nfs_client)) break; + set_bit(NFS_STATE_CHANGE_WAIT, &state->flags); prepare_to_wait(&state->waitq, &wait, TASK_KILLABLE); /* * Ensure we process the state changes in the same order @@ -1693,6 +1686,7 @@ static void nfs_set_open_stateid_locked(struct nfs4_state *state, spin_unlock(&state->owner->so_lock); rcu_read_unlock(); trace_nfs4_open_stateid_update_wait(state->inode, stateid, 0); + if (!signal_pending(current)) { if (schedule_timeout(5*HZ) == 0) status = -EAGAIN; @@ -3435,7 +3429,8 @@ static bool nfs4_refresh_open_old_stateid(nfs4_stateid *dst, __be32 seqid_open; u32 dst_seqid; bool ret; - int seq; + int seq, status = -EAGAIN; + DEFINE_WAIT(wait); for (;;) { ret = false; @@ -3447,15 +3442,41 @@ static bool nfs4_refresh_open_old_stateid(nfs4_stateid *dst, continue; break; } + + write_seqlock(&state->seqlock); seqid_open = state->open_stateid.seqid; - if (read_seqretry(&state->seqlock, seq)) - continue; dst_seqid = be32_to_cpu(dst->seqid); - if ((s32)(dst_seqid - be32_to_cpu(seqid_open)) >= 0) - dst->seqid = cpu_to_be32(dst_seqid + 1); - else + + /* Did another OPEN bump the state's seqid? try again: */ + if ((s32)(be32_to_cpu(seqid_open) - dst_seqid) > 0) { dst->seqid = seqid_open; + write_sequnlock(&state->seqlock); + ret = true; + break; + } + + /* server says we're behind but we haven't seen the update yet */ + set_bit(NFS_STATE_CHANGE_WAIT, &state->flags); + prepare_to_wait(&state->waitq, &wait, TASK_KILLABLE); + write_sequnlock(&state->seqlock); + trace_nfs4_close_stateid_update_wait(state->inode, dst, 0); + + if (signal_pending(current)) + status = -EINTR; + else + if (schedule_timeout(5*HZ) != 0) + status = 0; + + finish_wait(&state->waitq, &wait); + + if (!status) + continue; + if (status == -EINTR) + break; + + /* we slept the whole 5 seconds, we must have lost a seqid */ + dst->seqid = cpu_to_be32(dst_seqid + 1); ret = true; break; } @@ -3632,9 +3653,10 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data) if (calldata->arg.fmode == 0 || calldata->arg.fmode == FMODE_READ) { /* Close-to-open cache consistency revalidation */ - if (!nfs4_have_delegation(inode, FMODE_READ)) + if (!nfs4_have_delegation(inode, FMODE_READ)) { calldata->arg.bitmask = NFS_SERVER(inode)->cache_consistency_bitmask; - else + nfs4_bitmask_adjust(calldata->arg.bitmask, inode, NFS_SERVER(inode), NULL); + } else calldata->arg.bitmask = NULL; } @@ -5255,28 +5277,60 @@ static bool nfs4_read_stateid_changed(struct rpc_task *task, return true; } -static int nfs4_read_done(struct rpc_task *task, struct nfs_pgio_header *hdr) +static bool nfs4_read_plus_not_supported(struct rpc_task *task, + struct nfs_pgio_header *hdr) { + struct nfs_server *server = NFS_SERVER(hdr->inode); + struct rpc_message *msg = &task->tk_msg; + if (msg->rpc_proc == &nfs4_procedures[NFSPROC4_CLNT_READ_PLUS] && + server->caps & NFS_CAP_READ_PLUS && task->tk_status == -ENOTSUPP) { + server->caps &= ~NFS_CAP_READ_PLUS; + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ]; + rpc_restart_call_prepare(task); + return true; + } + return false; +} + +static int nfs4_read_done(struct rpc_task *task, struct nfs_pgio_header *hdr) +{ dprintk("--> %s\n", __func__); if (!nfs4_sequence_done(task, &hdr->res.seq_res)) return -EAGAIN; if (nfs4_read_stateid_changed(task, &hdr->args)) return -EAGAIN; + if (nfs4_read_plus_not_supported(task, hdr)) + return -EAGAIN; if (task->tk_status > 0) nfs_invalidate_atime(hdr->inode); return hdr->pgio_done_cb ? hdr->pgio_done_cb(task, hdr) : nfs4_read_done_cb(task, hdr); } +#ifdef CONFIG_NFS_V4_2 +static void nfs42_read_plus_support(struct nfs_server *server, struct rpc_message *msg) +{ + if (server->caps & NFS_CAP_READ_PLUS) + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ_PLUS]; + else + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ]; +} +#else +static void nfs42_read_plus_support(struct nfs_server *server, struct rpc_message *msg) +{ + msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ]; +} +#endif /* CONFIG_NFS_V4_2 */ + static void nfs4_proc_read_setup(struct nfs_pgio_header *hdr, struct rpc_message *msg) { hdr->timestamp = jiffies; if (!hdr->pgio_done_cb) hdr->pgio_done_cb = nfs4_read_done_cb; - msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ]; + nfs42_read_plus_support(NFS_SERVER(hdr->inode), msg); nfs4_init_sequence(&hdr->args.seq_args, &hdr->res.seq_res, 0, 0); } @@ -5360,6 +5414,38 @@ bool nfs4_write_need_cache_consistency_data(struct nfs_pgio_header *hdr) return nfs4_have_delegation(hdr->inode, FMODE_READ) == 0; } +static void nfs4_bitmask_adjust(__u32 *bitmask, struct inode *inode, + struct nfs_server *server, + struct nfs4_label *label) +{ + + unsigned long cache_validity = READ_ONCE(NFS_I(inode)->cache_validity); + + if ((cache_validity & NFS_INO_INVALID_DATA) || + (cache_validity & NFS_INO_REVAL_PAGECACHE) || + (cache_validity & NFS_INO_REVAL_FORCED) || + (cache_validity & NFS_INO_INVALID_OTHER)) + nfs4_bitmap_copy_adjust(bitmask, nfs4_bitmask(server, label), inode); + + if (cache_validity & NFS_INO_INVALID_ATIME) + bitmask[1] |= FATTR4_WORD1_TIME_ACCESS; + if (cache_validity & NFS_INO_INVALID_ACCESS) + bitmask[0] |= FATTR4_WORD1_MODE | FATTR4_WORD1_OWNER | + FATTR4_WORD1_OWNER_GROUP; + if (cache_validity & NFS_INO_INVALID_ACL) + bitmask[0] |= FATTR4_WORD0_ACL; + if (cache_validity & NFS_INO_INVALID_LABEL) + bitmask[2] |= FATTR4_WORD2_SECURITY_LABEL; + if (cache_validity & NFS_INO_INVALID_CTIME) + bitmask[0] |= FATTR4_WORD0_CHANGE; + if (cache_validity & NFS_INO_INVALID_MTIME) + bitmask[1] |= FATTR4_WORD1_TIME_MODIFY; + if (cache_validity & NFS_INO_INVALID_SIZE) + bitmask[0] |= FATTR4_WORD0_SIZE; + if (cache_validity & NFS_INO_INVALID_BLOCKS) + bitmask[1] |= FATTR4_WORD1_SPACE_USED; +} + static void nfs4_proc_write_setup(struct nfs_pgio_header *hdr, struct rpc_message *msg, struct rpc_clnt **clnt) @@ -5369,8 +5455,10 @@ static void nfs4_proc_write_setup(struct nfs_pgio_header *hdr, if (!nfs4_write_need_cache_consistency_data(hdr)) { hdr->args.bitmask = NULL; hdr->res.fattr = NULL; - } else + } else { hdr->args.bitmask = server->cache_consistency_bitmask; + nfs4_bitmask_adjust(hdr->args.bitmask, hdr->inode, server, NULL); + } if (!hdr->pgio_done_cb) hdr->pgio_done_cb = nfs4_write_done_cb; @@ -6006,9 +6094,34 @@ static void nfs4_init_boot_verifier(const struct nfs_client *clp, memcpy(bootverf->data, verf, sizeof(bootverf->data)); } +static size_t +nfs4_get_uniquifier(struct nfs_client *clp, char *buf, size_t buflen) +{ + struct nfs_net *nn = net_generic(clp->cl_net, nfs_net_id); + struct nfs_netns_client *nn_clp = nn->nfs_client; + const char *id; + + buf[0] = '\0'; + + if (nn_clp) { + rcu_read_lock(); + id = rcu_dereference(nn_clp->identifier); + if (id) + strscpy(buf, id, buflen); + rcu_read_unlock(); + } + + if (nfs4_client_id_uniquifier[0] != '\0' && buf[0] == '\0') + strscpy(buf, nfs4_client_id_uniquifier, buflen); + + return strlen(buf); +} + static int nfs4_init_nonuniform_client_string(struct nfs_client *clp) { + char buf[NFS4_CLIENT_ID_UNIQ_LEN]; + size_t buflen; size_t len; char *str; @@ -6022,8 +6135,11 @@ nfs4_init_nonuniform_client_string(struct nfs_client *clp) strlen(rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR)) + 1; rcu_read_unlock(); - if (nfs4_client_id_uniquifier[0] != '\0') - len += strlen(nfs4_client_id_uniquifier) + 1; + + buflen = nfs4_get_uniquifier(clp, buf, sizeof(buf)); + if (buflen) + len += buflen + 1; + if (len > NFS4_OPAQUE_LIMIT + 1) return -EINVAL; @@ -6037,10 +6153,9 @@ nfs4_init_nonuniform_client_string(struct nfs_client *clp) return -ENOMEM; rcu_read_lock(); - if (nfs4_client_id_uniquifier[0] != '\0') + if (buflen) scnprintf(str, len, "Linux NFSv4.0 %s/%s/%s", - clp->cl_rpcclient->cl_nodename, - nfs4_client_id_uniquifier, + clp->cl_rpcclient->cl_nodename, buf, rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR)); else @@ -6055,50 +6170,23 @@ nfs4_init_nonuniform_client_string(struct nfs_client *clp) } static int -nfs4_init_uniquifier_client_string(struct nfs_client *clp) -{ - size_t len; - char *str; - - len = 10 + 10 + 1 + 10 + 1 + - strlen(nfs4_client_id_uniquifier) + 1 + - strlen(clp->cl_rpcclient->cl_nodename) + 1; - - if (len > NFS4_OPAQUE_LIMIT + 1) - return -EINVAL; - - /* - * Since this string is allocated at mount time, and held until the - * nfs_client is destroyed, we can use GFP_KERNEL here w/o worrying - * about a memory-reclaim deadlock. - */ - str = kmalloc(len, GFP_KERNEL); - if (!str) - return -ENOMEM; - - scnprintf(str, len, "Linux NFSv%u.%u %s/%s", - clp->rpc_ops->version, clp->cl_minorversion, - nfs4_client_id_uniquifier, - clp->cl_rpcclient->cl_nodename); - clp->cl_owner_id = str; - return 0; -} - -static int nfs4_init_uniform_client_string(struct nfs_client *clp) { + char buf[NFS4_CLIENT_ID_UNIQ_LEN]; + size_t buflen; size_t len; char *str; if (clp->cl_owner_id != NULL) return 0; - if (nfs4_client_id_uniquifier[0] != '\0') - return nfs4_init_uniquifier_client_string(clp); - len = 10 + 10 + 1 + 10 + 1 + strlen(clp->cl_rpcclient->cl_nodename) + 1; + buflen = nfs4_get_uniquifier(clp, buf, sizeof(buf)); + if (buflen) + len += buflen + 1; + if (len > NFS4_OPAQUE_LIMIT + 1) return -EINVAL; @@ -6111,9 +6199,14 @@ nfs4_init_uniform_client_string(struct nfs_client *clp) if (!str) return -ENOMEM; - scnprintf(str, len, "Linux NFSv%u.%u %s", - clp->rpc_ops->version, clp->cl_minorversion, - clp->cl_rpcclient->cl_nodename); + if (buflen) + scnprintf(str, len, "Linux NFSv%u.%u %s/%s", + clp->rpc_ops->version, clp->cl_minorversion, + buf, clp->cl_rpcclient->cl_nodename); + else + scnprintf(str, len, "Linux NFSv%u.%u %s", + clp->rpc_ops->version, clp->cl_minorversion, + clp->cl_rpcclient->cl_nodename); clp->cl_owner_id = str; return 0; } @@ -6406,6 +6499,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, const struct cred *cred, data->args.fhandle = &data->fh; data->args.stateid = &data->stateid; data->args.bitmask = server->cache_consistency_bitmask; + nfs4_bitmask_adjust(data->args.bitmask, inode, server, NULL); nfs_copy_fh(&data->fh, NFS_FH(inode)); nfs4_stateid_copy(&data->stateid, stateid); data->res.fattr = &data->fattr; @@ -7440,7 +7534,7 @@ nfs4_listxattr_nfs4_label(struct inode *inode, char *list, size_t list_len) if (nfs_server_capable(inode, NFS_CAP_SECURITY_LABEL)) { len = security_inode_listsecurity(inode, list, list_len); - if (list_len && len > list_len) + if (len >= 0 && list_len && len > list_len) return -ERANGE; } return len; @@ -8039,9 +8133,11 @@ int nfs4_proc_secinfo(struct inode *dir, const struct qstr *name, * both PNFS and NON_PNFS flags set, and not having one of NON_PNFS, PNFS, or * DS flags set. */ -static int nfs4_check_cl_exchange_flags(u32 flags) +static int nfs4_check_cl_exchange_flags(u32 flags, u32 version) { - if (flags & ~EXCHGID4_FLAG_MASK_R) + if (version >= 2 && (flags & ~EXCHGID4_2_FLAG_MASK_R)) + goto out_inval; + else if (version < 2 && (flags & ~EXCHGID4_FLAG_MASK_R)) goto out_inval; if ((flags & EXCHGID4_FLAG_USE_PNFS_MDS) && (flags & EXCHGID4_FLAG_USE_NON_PNFS)) @@ -8454,7 +8550,8 @@ static int _nfs4_proc_exchange_id(struct nfs_client *clp, const struct cred *cre if (status != 0) goto out; - status = nfs4_check_cl_exchange_flags(resp->flags); + status = nfs4_check_cl_exchange_flags(resp->flags, + clp->cl_mvops->minor_version); if (status != 0) goto out; @@ -9693,7 +9790,6 @@ _nfs41_proc_secinfo_no_name(struct nfs_server *server, struct nfs_fh *fhandle, .rpc_argp = &args, .rpc_resp = &res, }; - struct rpc_clnt *clnt = server->client; struct nfs4_call_sync_data data = { .seq_server = server, .seq_args = &args.seq_args, @@ -9710,8 +9806,7 @@ _nfs41_proc_secinfo_no_name(struct nfs_server *server, struct nfs_fh *fhandle, int status; if (use_integrity) { - clnt = server->nfs_client->cl_rpcclient; - task_setup.rpc_client = clnt; + task_setup.rpc_client = server->nfs_client->cl_rpcclient; cred = nfs4_get_clid_cred(server->nfs_client); msg.rpc_cred = cred; @@ -10165,7 +10260,8 @@ static const struct nfs4_minor_version_ops nfs_v4_2_minor_ops = { | NFS_CAP_SEEK | NFS_CAP_LAYOUTSTATS | NFS_CAP_CLONE - | NFS_CAP_LAYOUTERROR, + | NFS_CAP_LAYOUTERROR + | NFS_CAP_READ_PLUS, .init_client = nfs41_init_client, .shutdown_client = nfs41_shutdown_client, .match_stateid = nfs41_match_stateid, diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index b4f852d4d099..484c1da96dea 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -1511,6 +1511,7 @@ DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_setattr); DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_delegreturn); DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_open_stateid_update); DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_open_stateid_update_wait); +DEFINE_NFS4_INODE_STATEID_EVENT(nfs4_close_stateid_update_wait); DECLARE_EVENT_CLASS(nfs4_getattr_event, TP_PROTO( diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 0b3510f62623..c6dbfcae7517 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -5308,7 +5308,6 @@ static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, uint32_t attrlen, bitmap[3] = {0}; int status; - unsigned int pg_offset; res->acl_len = 0; if ((status = decode_op_hdr(xdr, OP_GETATTR)) != 0) @@ -5316,9 +5315,6 @@ static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, xdr_enter_page(xdr, xdr->buf->page_len); - /* Calculate the offset of the page data */ - pg_offset = xdr->buf->head[0].iov_len; - if ((status = decode_attr_bitmap(xdr, bitmap)) != 0) goto out; if ((status = decode_attr_length(xdr, &attrlen, &savep)) != 0) @@ -5331,7 +5327,7 @@ static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, /* The bitmap (xdr len + bitmaps) and the attr xdr len words * are stored with the acl data to handle the problem of * variable length bitmaps.*/ - res->acl_data_offset = xdr_stream_pos(xdr) - pg_offset; + res->acl_data_offset = xdr_page_pos(xdr); res->acl_len = attrlen; /* Check for receive buffer overflow */ @@ -7619,6 +7615,7 @@ const struct rpc_procinfo nfs4_procedures[] = { PROC42(SETXATTR, enc_setxattr, dec_setxattr), PROC42(LISTXATTRS, enc_listxattrs, dec_listxattrs), PROC42(REMOVEXATTR, enc_removexattr, dec_removexattr), + PROC42(READ_PLUS, enc_read_plus, dec_read_plus), }; static unsigned int nfs_version4_counts[ARRAY_SIZE(nfs4_procedures)]; diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 71f7741126b6..0e50b9d45c32 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -902,7 +902,7 @@ restart: } /* - * Called by the state manger to remove all layouts established under an + * Called by the state manager to remove all layouts established under an * expired lease. */ void diff --git a/fs/nfs/super.c b/fs/nfs/super.c index f943e37853fa..78c46a517fcf 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -889,7 +889,7 @@ static struct nfs_server *nfs_try_mount_request(struct fs_context *fc) default: if (rpcauth_get_gssinfo(flavor, &info) != 0) continue; - /* Fallthrough */ + break; } dfprintk(MOUNT, "NFS: attempting to use auth flavor %u\n", flavor); ctx->selected_flavor = flavor; diff --git a/fs/nfs/sysfs.c b/fs/nfs/sysfs.c index c489496b5659..8cb70755e3c9 100644 --- a/fs/nfs/sysfs.c +++ b/fs/nfs/sysfs.c @@ -79,7 +79,12 @@ static ssize_t nfs_netns_identifier_show(struct kobject *kobj, struct nfs_netns_client *c = container_of(kobj, struct nfs_netns_client, kobject); - return scnprintf(buf, PAGE_SIZE, "%s\n", c->identifier); + ssize_t ret; + + rcu_read_lock(); + ret = scnprintf(buf, PAGE_SIZE, "%s\n", rcu_dereference(c->identifier)); + rcu_read_unlock(); + return ret; } /* Strip trailing '\n' */ @@ -107,7 +112,7 @@ static ssize_t nfs_netns_identifier_store(struct kobject *kobj, p = kmemdup_nul(buf, len, GFP_KERNEL); if (!p) return -ENOMEM; - old = xchg(&c->identifier, p); + old = rcu_dereference_protected(xchg(&c->identifier, (char __rcu *)p), 1); if (old) { synchronize_rcu(); kfree(old); @@ -121,7 +126,7 @@ static void nfs_netns_client_release(struct kobject *kobj) struct nfs_netns_client, kobject); - kfree(c->identifier); + kfree(rcu_dereference_raw(c->identifier)); kfree(c); } diff --git a/fs/nfs/sysfs.h b/fs/nfs/sysfs.h index ebcbdc40483b..5501ef573c32 100644 --- a/fs/nfs/sysfs.h +++ b/fs/nfs/sysfs.h @@ -11,7 +11,7 @@ struct nfs_netns_client { struct kobject kobject; struct net *net; - const char *identifier; + const char __rcu *identifier; }; extern struct kobject *nfs_client_kobj; diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index b8360be141da..9dc7eeac924f 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -551,13 +551,13 @@ enum { NFSPROC4_CLNT_LOOKUPP, NFSPROC4_CLNT_LAYOUTERROR, - NFSPROC4_CLNT_COPY_NOTIFY, NFSPROC4_CLNT_GETXATTR, NFSPROC4_CLNT_SETXATTR, NFSPROC4_CLNT_LISTXATTRS, NFSPROC4_CLNT_REMOVEXATTR, + NFSPROC4_CLNT_READ_PLUS, }; /* nfs41 types */ diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index 7eae72a8762e..38e60ec742df 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -287,5 +287,6 @@ struct nfs_server { #define NFS_CAP_LAYOUTERROR (1U << 26) #define NFS_CAP_COPY_NOTIFY (1U << 27) #define NFS_CAP_XATTR (1U << 28) +#define NFS_CAP_READ_PLUS (1U << 29) #endif diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 69cb46f7b8d2..d63cb862d58e 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -525,7 +525,7 @@ struct nfs_closeargs { struct nfs_seqid * seqid; fmode_t fmode; u32 share_access; - const u32 * bitmask; + u32 * bitmask; struct nfs4_layoutreturn_args *lr_args; }; @@ -608,7 +608,7 @@ struct nfs4_delegreturnargs { struct nfs4_sequence_args seq_args; const struct nfs_fh *fhandle; const nfs4_stateid *stateid; - const u32 * bitmask; + u32 * bitmask; struct nfs4_layoutreturn_args *lr_args; }; @@ -648,7 +648,7 @@ struct nfs_pgio_args { union { unsigned int replen; /* used by read */ struct { - const u32 * bitmask; /* used by write */ + u32 * bitmask; /* used by write */ enum nfs3_stable_how stable; /* used by write */ }; }; @@ -657,7 +657,7 @@ struct nfs_pgio_args { struct nfs_pgio_res { struct nfs4_sequence_res seq_res; struct nfs_fattr * fattr; - __u32 count; + __u64 count; __u32 op_status; union { struct { diff --git a/include/linux/sunrpc/bc_xprt.h b/include/linux/sunrpc/bc_xprt.h index d796058cdff2..f07c334c599f 100644 --- a/include/linux/sunrpc/bc_xprt.h +++ b/include/linux/sunrpc/bc_xprt.h @@ -4,7 +4,7 @@ NetApp provides this source code under the GPL v2 License. The GPL v2 license is available at -http://opensource.org/licenses/gpl-license.php. +https://opensource.org/licenses/gpl-license.php. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT diff --git a/include/linux/sunrpc/cache.h b/include/linux/sunrpc/cache.h index 10891b70fc7b..d0965e2997b0 100644 --- a/include/linux/sunrpc/cache.h +++ b/include/linux/sunrpc/cache.h @@ -45,7 +45,8 @@ */ struct cache_head { struct hlist_node cache_list; - time64_t expiry_time; /* After time time, don't use the data */ + time64_t expiry_time; /* After time expiry_time, don't use + * the data */ time64_t last_refresh; /* If CACHE_PENDING, this is when upcall was * sent, else this is when update was * received, though it is alway set to diff --git a/include/linux/sunrpc/msg_prot.h b/include/linux/sunrpc/msg_prot.h index bea40d9f03a1..43f854487539 100644 --- a/include/linux/sunrpc/msg_prot.h +++ b/include/linux/sunrpc/msg_prot.h @@ -143,7 +143,7 @@ typedef __be32 rpc_fraghdr; /* * Well-known netids. See: * - * http://www.iana.org/assignments/rpc-netids/rpc-netids.xhtml + * https://www.iana.org/assignments/rpc-netids/rpc-netids.xhtml */ #define RPCBIND_NETID_UDP "udp" #define RPCBIND_NETID_TCP "tcp" diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 5a6a81b7cd9f..fe7ff7f5b584 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -240,6 +240,7 @@ extern int xdr_restrict_buflen(struct xdr_stream *xdr, int newbuflen); extern void xdr_write_pages(struct xdr_stream *xdr, struct page **pages, unsigned int base, unsigned int len); extern unsigned int xdr_stream_pos(const struct xdr_stream *xdr); +extern unsigned int xdr_page_pos(const struct xdr_stream *xdr); extern void xdr_init_decode(struct xdr_stream *xdr, struct xdr_buf *buf, __be32 *p, struct rpc_rqst *rqst); extern void xdr_init_decode_pages(struct xdr_stream *xdr, struct xdr_buf *buf, @@ -249,6 +250,8 @@ extern __be32 *xdr_inline_decode(struct xdr_stream *xdr, size_t nbytes); extern unsigned int xdr_read_pages(struct xdr_stream *xdr, unsigned int len); extern void xdr_enter_page(struct xdr_stream *xdr, unsigned int len); extern int xdr_process_buf(struct xdr_buf *buf, unsigned int offset, unsigned int len, int (*actor)(struct scatterlist *, void *), void *data); +extern uint64_t xdr_align_data(struct xdr_stream *, uint64_t, uint32_t); +extern uint64_t xdr_expand_hole(struct xdr_stream *, uint64_t, uint64_t); /** * xdr_stream_remaining - Return the number of bytes remaining in the stream diff --git a/include/trace/events/rpcrdma.h b/include/trace/events/rpcrdma.h index b6aad52beb62..bf1065772228 100644 --- a/include/trace/events/rpcrdma.h +++ b/include/trace/events/rpcrdma.h @@ -424,7 +424,6 @@ DEFINE_CONN_EVENT(connect); DEFINE_CONN_EVENT(disconnect); DEFINE_RXPRT_EVENT(xprtrdma_op_inject_dsc); -DEFINE_RXPRT_EVENT(xprtrdma_op_setport); TRACE_EVENT(xprtrdma_op_connect, TP_PROTO( @@ -1189,68 +1188,6 @@ TRACE_EVENT(xprtrdma_decode_seg, ); /** - ** Allocation/release of rpcrdma_reqs and rpcrdma_reps - **/ - -TRACE_EVENT(xprtrdma_op_allocate, - TP_PROTO( - const struct rpc_task *task, - const struct rpcrdma_req *req - ), - - TP_ARGS(task, req), - - TP_STRUCT__entry( - __field(unsigned int, task_id) - __field(unsigned int, client_id) - __field(const void *, req) - __field(size_t, callsize) - __field(size_t, rcvsize) - ), - - TP_fast_assign( - __entry->task_id = task->tk_pid; - __entry->client_id = task->tk_client->cl_clid; - __entry->req = req; - __entry->callsize = task->tk_rqstp->rq_callsize; - __entry->rcvsize = task->tk_rqstp->rq_rcvsize; - ), - - TP_printk("task:%u@%u req=%p (%zu, %zu)", - __entry->task_id, __entry->client_id, - __entry->req, __entry->callsize, __entry->rcvsize - ) -); - -TRACE_EVENT(xprtrdma_op_free, - TP_PROTO( - const struct rpc_task *task, - const struct rpcrdma_req *req - ), - - TP_ARGS(task, req), - - TP_STRUCT__entry( - __field(unsigned int, task_id) - __field(unsigned int, client_id) - __field(const void *, req) - __field(const void *, rep) - ), - - TP_fast_assign( - __entry->task_id = task->tk_pid; - __entry->client_id = task->tk_client->cl_clid; - __entry->req = req; - __entry->rep = req->rl_reply; - ), - - TP_printk("task:%u@%u req=%p rep=%p", - __entry->task_id, __entry->client_id, - __entry->req, __entry->rep - ) -); - -/** ** Callback events **/ diff --git a/include/trace/events/sunrpc.h b/include/trace/events/sunrpc.h index 65d7dfbbc9cd..f45b3c01370c 100644 --- a/include/trace/events/sunrpc.h +++ b/include/trace/events/sunrpc.h @@ -259,8 +259,10 @@ DECLARE_EVENT_CLASS(rpc_task_status, TP_ARGS(task)) DEFINE_RPC_STATUS_EVENT(call); -DEFINE_RPC_STATUS_EVENT(bind); DEFINE_RPC_STATUS_EVENT(connect); +DEFINE_RPC_STATUS_EVENT(timeout); +DEFINE_RPC_STATUS_EVENT(retry_refresh); +DEFINE_RPC_STATUS_EVENT(refresh); TRACE_EVENT(rpc_request, TP_PROTO(const struct rpc_task *task), @@ -385,7 +387,10 @@ DECLARE_EVENT_CLASS(rpc_task_running, DEFINE_RPC_RUNNING_EVENT(begin); DEFINE_RPC_RUNNING_EVENT(run_action); +DEFINE_RPC_RUNNING_EVENT(sync_sleep); +DEFINE_RPC_RUNNING_EVENT(sync_wake); DEFINE_RPC_RUNNING_EVENT(complete); +DEFINE_RPC_RUNNING_EVENT(timeout); DEFINE_RPC_RUNNING_EVENT(signalled); DEFINE_RPC_RUNNING_EVENT(end); @@ -517,6 +522,49 @@ DEFINE_RPC_REPLY_EVENT(stale_creds); DEFINE_RPC_REPLY_EVENT(bad_creds); DEFINE_RPC_REPLY_EVENT(auth_tooweak); +#define DEFINE_RPCB_ERROR_EVENT(name) \ + DEFINE_EVENT(rpc_reply_event, rpcb_##name##_err, \ + TP_PROTO( \ + const struct rpc_task *task \ + ), \ + TP_ARGS(task)) + +DEFINE_RPCB_ERROR_EVENT(prog_unavail); +DEFINE_RPCB_ERROR_EVENT(timeout); +DEFINE_RPCB_ERROR_EVENT(bind_version); +DEFINE_RPCB_ERROR_EVENT(unreachable); +DEFINE_RPCB_ERROR_EVENT(unrecognized); + +TRACE_EVENT(rpc_buf_alloc, + TP_PROTO( + const struct rpc_task *task, + int status + ), + + TP_ARGS(task, status), + + TP_STRUCT__entry( + __field(unsigned int, task_id) + __field(unsigned int, client_id) + __field(size_t, callsize) + __field(size_t, recvsize) + __field(int, status) + ), + + TP_fast_assign( + __entry->task_id = task->tk_pid; + __entry->client_id = task->tk_client->cl_clid; + __entry->callsize = task->tk_rqstp->rq_callsize; + __entry->recvsize = task->tk_rqstp->rq_rcvsize; + __entry->status = status; + ), + + TP_printk("task:%u@%u callsize=%zu recvsize=%zu status=%d", + __entry->task_id, __entry->client_id, + __entry->callsize, __entry->recvsize, __entry->status + ) +); + TRACE_EVENT(rpc_call_rpcerror, TP_PROTO( const struct rpc_task *task, @@ -868,6 +916,34 @@ DEFINE_RPC_SOCKET_EVENT_DONE(rpc_socket_reset_connection); DEFINE_RPC_SOCKET_EVENT(rpc_socket_close); DEFINE_RPC_SOCKET_EVENT(rpc_socket_shutdown); +TRACE_EVENT(rpc_socket_nospace, + TP_PROTO( + const struct rpc_rqst *rqst, + const struct sock_xprt *transport + ), + + TP_ARGS(rqst, transport), + + TP_STRUCT__entry( + __field(unsigned int, task_id) + __field(unsigned int, client_id) + __field(unsigned int, total) + __field(unsigned int, remaining) + ), + + TP_fast_assign( + __entry->task_id = rqst->rq_task->tk_pid; + __entry->client_id = rqst->rq_task->tk_client->cl_clid; + __entry->total = rqst->rq_slen; + __entry->remaining = rqst->rq_slen - transport->xmit.offset; + ), + + TP_printk("task:%u@%u total=%u remaining=%u", + __entry->task_id, __entry->client_id, + __entry->total, __entry->remaining + ) +); + TRACE_DEFINE_ENUM(XPRT_LOCKED); TRACE_DEFINE_ENUM(XPRT_CONNECTED); TRACE_DEFINE_ENUM(XPRT_CONNECTING); @@ -925,6 +1001,7 @@ DECLARE_EVENT_CLASS(rpc_xprt_lifetime_class, TP_ARGS(xprt)) DEFINE_RPC_XPRT_LIFETIME_EVENT(create); +DEFINE_RPC_XPRT_LIFETIME_EVENT(connect); DEFINE_RPC_XPRT_LIFETIME_EVENT(disconnect_auto); DEFINE_RPC_XPRT_LIFETIME_EVENT(disconnect_done); DEFINE_RPC_XPRT_LIFETIME_EVENT(disconnect_force); @@ -969,7 +1046,6 @@ DECLARE_EVENT_CLASS(rpc_xprt_event, DEFINE_RPC_XPRT_EVENT(timer); DEFINE_RPC_XPRT_EVENT(lookup_rqst); -DEFINE_RPC_XPRT_EVENT(complete_rqst); TRACE_EVENT(xprt_transmit, TP_PROTO( @@ -1002,37 +1078,6 @@ TRACE_EVENT(xprt_transmit, __entry->seqno, __entry->status) ); -TRACE_EVENT(xprt_enq_xmit, - TP_PROTO( - const struct rpc_task *task, - int stage - ), - - TP_ARGS(task, stage), - - TP_STRUCT__entry( - __field(unsigned int, task_id) - __field(unsigned int, client_id) - __field(u32, xid) - __field(u32, seqno) - __field(int, stage) - ), - - TP_fast_assign( - __entry->task_id = task->tk_pid; - __entry->client_id = task->tk_client ? - task->tk_client->cl_clid : -1; - __entry->xid = be32_to_cpu(task->tk_rqstp->rq_xid); - __entry->seqno = task->tk_rqstp->rq_seqno; - __entry->stage = stage; - ), - - TP_printk( - "task:%u@%u xid=0x%08x seqno=%u stage=%d", - __entry->task_id, __entry->client_id, __entry->xid, - __entry->seqno, __entry->stage) -); - TRACE_EVENT(xprt_ping, TP_PROTO(const struct rpc_xprt *xprt, int status), @@ -1095,6 +1140,7 @@ DECLARE_EVENT_CLASS(xprt_writelock_event, DEFINE_WRITELOCK_EVENT(reserve_xprt); DEFINE_WRITELOCK_EVENT(release_xprt); +DEFINE_WRITELOCK_EVENT(transmit_queued); DECLARE_EVENT_CLASS(xprt_cong_event, TP_PROTO( @@ -1147,6 +1193,30 @@ DEFINE_CONG_EVENT(release_cong); DEFINE_CONG_EVENT(get_cong); DEFINE_CONG_EVENT(put_cong); +TRACE_EVENT(xprt_reserve, + TP_PROTO( + const struct rpc_rqst *rqst + ), + + TP_ARGS(rqst), + + TP_STRUCT__entry( + __field(unsigned int, task_id) + __field(unsigned int, client_id) + __field(u32, xid) + ), + + TP_fast_assign( + __entry->task_id = rqst->rq_task->tk_pid; + __entry->client_id = rqst->rq_task->tk_client->cl_clid; + __entry->xid = be32_to_cpu(rqst->rq_xid); + ), + + TP_printk("task:%u@%u xid=0x%08x", + __entry->task_id, __entry->client_id, __entry->xid + ) +); + TRACE_EVENT(xs_stream_read_data, TP_PROTO(struct rpc_xprt *xprt, ssize_t err, size_t total), @@ -1202,6 +1272,156 @@ TRACE_EVENT(xs_stream_read_request, __entry->copied, __entry->reclen, __entry->offset) ); +TRACE_EVENT(rpcb_getport, + TP_PROTO( + const struct rpc_clnt *clnt, + const struct rpc_task *task, + unsigned int bind_version + ), + + TP_ARGS(clnt, task, bind_version), + + TP_STRUCT__entry( + __field(unsigned int, task_id) + __field(unsigned int, client_id) + __field(unsigned int, program) + __field(unsigned int, version) + __field(int, protocol) + __field(unsigned int, bind_version) + __string(servername, task->tk_xprt->servername) + ), + + TP_fast_assign( + __entry->task_id = task->tk_pid; + __entry->client_id = clnt->cl_clid; + __entry->program = clnt->cl_prog; + __entry->version = clnt->cl_vers; + __entry->protocol = task->tk_xprt->prot; + __entry->bind_version = bind_version; + __assign_str(servername, task->tk_xprt->servername); + ), + + TP_printk("task:%u@%u server=%s program=%u version=%u protocol=%d bind_version=%u", + __entry->task_id, __entry->client_id, __get_str(servername), + __entry->program, __entry->version, __entry->protocol, + __entry->bind_version + ) +); + +TRACE_EVENT(rpcb_setport, + TP_PROTO( + const struct rpc_task *task, + int status, + unsigned short port + ), + + TP_ARGS(task, status, port), + + TP_STRUCT__entry( + __field(unsigned int, task_id) + __field(unsigned int, client_id) + __field(int, status) + __field(unsigned short, port) + ), + + TP_fast_assign( + __entry->task_id = task->tk_pid; + __entry->client_id = task->tk_client->cl_clid; + __entry->status = status; + __entry->port = port; + ), + + TP_printk("task:%u@%u status=%d port=%u", + __entry->task_id, __entry->client_id, + __entry->status, __entry->port + ) +); + +TRACE_EVENT(pmap_register, + TP_PROTO( + u32 program, + u32 version, + int protocol, + unsigned short port + ), + + TP_ARGS(program, version, protocol, port), + + TP_STRUCT__entry( + __field(unsigned int, program) + __field(unsigned int, version) + __field(int, protocol) + __field(unsigned int, port) + ), + + TP_fast_assign( + __entry->program = program; + __entry->version = version; + __entry->protocol = protocol; + __entry->port = port; + ), + + TP_printk("program=%u version=%u protocol=%d port=%u", + __entry->program, __entry->version, + __entry->protocol, __entry->port + ) +); + +TRACE_EVENT(rpcb_register, + TP_PROTO( + u32 program, + u32 version, + const char *addr, + const char *netid + ), + + TP_ARGS(program, version, addr, netid), + + TP_STRUCT__entry( + __field(unsigned int, program) + __field(unsigned int, version) + __string(addr, addr) + __string(netid, netid) + ), + + TP_fast_assign( + __entry->program = program; + __entry->version = version; + __assign_str(addr, addr); + __assign_str(netid, netid); + ), + + TP_printk("program=%u version=%u addr=%s netid=%s", + __entry->program, __entry->version, + __get_str(addr), __get_str(netid) + ) +); + +TRACE_EVENT(rpcb_unregister, + TP_PROTO( + u32 program, + u32 version, + const char *netid + ), + + TP_ARGS(program, version, netid), + + TP_STRUCT__entry( + __field(unsigned int, program) + __field(unsigned int, version) + __string(netid, netid) + ), + + TP_fast_assign( + __entry->program = program; + __entry->version = version; + __assign_str(netid, netid); + ), + + TP_printk("program=%u version=%u netid=%s", + __entry->program, __entry->version, __get_str(netid) + ) +); DECLARE_EVENT_CLASS(svc_xdr_buf_class, TP_PROTO( diff --git a/include/uapi/linux/nfs4.h b/include/uapi/linux/nfs4.h index bf197e99b98f..ed5415e0f1c1 100644 --- a/include/uapi/linux/nfs4.h +++ b/include/uapi/linux/nfs4.h @@ -139,6 +139,8 @@ #define EXCHGID4_FLAG_UPD_CONFIRMED_REC_A 0x40000000 #define EXCHGID4_FLAG_CONFIRMED_R 0x80000000 + +#define EXCHGID4_FLAG_SUPP_FENCE_OPS 0x00000004 /* * Since the validity of these bits depends on whether * they're set in the argument or response, have separate @@ -146,6 +148,7 @@ */ #define EXCHGID4_FLAG_MASK_A 0x40070103 #define EXCHGID4_FLAG_MASK_R 0x80070103 +#define EXCHGID4_2_FLAG_MASK_R 0x80070107 #define SEQ4_STATUS_CB_PATH_DOWN 0x00000001 #define SEQ4_STATUS_CB_GSS_CONTEXTS_EXPIRING 0x00000002 diff --git a/net/sunrpc/backchannel_rqst.c b/net/sunrpc/backchannel_rqst.c index 195b40c5dae4..3fecad369592 100644 --- a/net/sunrpc/backchannel_rqst.c +++ b/net/sunrpc/backchannel_rqst.c @@ -5,7 +5,7 @@ NetApp provides this source code under the GPL v2 License. The GPL v2 license is available at -http://opensource.org/licenses/gpl-license.php. +https://opensource.org/licenses/gpl-license.php. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 62e0b6c1e8cf..3259120462ed 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -47,10 +47,6 @@ # define RPCDBG_FACILITY RPCDBG_CALL #endif -#define dprint_status(t) \ - dprintk("RPC: %5u %s (status %d)\n", t->tk_pid, \ - __func__, t->tk_status) - /* * All RPC clients are linked into this list */ @@ -1639,10 +1635,6 @@ call_start(struct rpc_task *task) int idx = task->tk_msg.rpc_proc->p_statidx; trace_rpc_request(task); - dprintk("RPC: %5u call_start %s%d proc %s (%s)\n", task->tk_pid, - clnt->cl_program->name, clnt->cl_vers, - rpc_proc_name(task), - (RPC_IS_ASYNC(task) ? "async" : "sync")); /* Increment call count (version might not be valid for ping) */ if (clnt->cl_program->version[clnt->cl_vers]) @@ -1658,8 +1650,6 @@ call_start(struct rpc_task *task) static void call_reserve(struct rpc_task *task) { - dprint_status(task); - task->tk_status = 0; task->tk_action = call_reserveresult; xprt_reserve(task); @@ -1675,8 +1665,6 @@ call_reserveresult(struct rpc_task *task) { int status = task->tk_status; - dprint_status(task); - /* * After a call to xprt_reserve(), we must have either * a request slot or else an error status. @@ -1717,8 +1705,6 @@ call_reserveresult(struct rpc_task *task) static void call_retry_reserve(struct rpc_task *task) { - dprint_status(task); - task->tk_status = 0; task->tk_action = call_reserveresult; xprt_retry_reserve(task); @@ -1730,8 +1716,6 @@ call_retry_reserve(struct rpc_task *task) static void call_refresh(struct rpc_task *task) { - dprint_status(task); - task->tk_action = call_refreshresult; task->tk_status = 0; task->tk_client->cl_stats->rpcauthrefresh++; @@ -1746,8 +1730,6 @@ call_refreshresult(struct rpc_task *task) { int status = task->tk_status; - dprint_status(task); - task->tk_status = 0; task->tk_action = call_refresh; switch (status) { @@ -1770,12 +1752,10 @@ call_refreshresult(struct rpc_task *task) if (!task->tk_cred_retry) break; task->tk_cred_retry--; - dprintk("RPC: %5u %s: retry refresh creds\n", - task->tk_pid, __func__); + trace_rpc_retry_refresh_status(task); return; } - dprintk("RPC: %5u %s: refresh creds failed with error %d\n", - task->tk_pid, __func__, status); + trace_rpc_refresh_status(task); rpc_call_rpcerror(task, status); } @@ -1792,8 +1772,6 @@ call_allocate(struct rpc_task *task) const struct rpc_procinfo *proc = task->tk_msg.rpc_proc; int status; - dprint_status(task); - task->tk_status = 0; task->tk_action = call_encode; @@ -1823,6 +1801,7 @@ call_allocate(struct rpc_task *task) req->rq_rcvsize <<= 2; status = xprt->ops->buf_alloc(task); + trace_rpc_buf_alloc(task, status); xprt_inject_disconnect(xprt); if (status == 0) return; @@ -1831,8 +1810,6 @@ call_allocate(struct rpc_task *task) return; } - dprintk("RPC: %5u rpc_buffer allocation failed\n", task->tk_pid); - if (RPC_IS_ASYNC(task) || !fatal_signal_pending(current)) { task->tk_action = call_allocate; rpc_delay(task, HZ>>4); @@ -1883,7 +1860,7 @@ call_encode(struct rpc_task *task) { if (!rpc_task_need_encode(task)) goto out; - dprint_status(task); + /* Dequeue task from the receive queue while we're encoding */ xprt_request_dequeue_xprt(task); /* Encode here so that rpcsec_gss can use correct sequence number. */ @@ -1902,8 +1879,7 @@ call_encode(struct rpc_task *task) } else { task->tk_action = call_refresh; task->tk_cred_retry--; - dprintk("RPC: %5u %s: retry refresh creds\n", - task->tk_pid, __func__); + trace_rpc_retry_refresh_status(task); } break; default: @@ -1960,8 +1936,6 @@ call_bind(struct rpc_task *task) return; } - dprint_status(task); - task->tk_action = call_bind_status; if (!xprt_prepare_transmit(task)) return; @@ -1983,8 +1957,6 @@ call_bind_status(struct rpc_task *task) return; } - dprint_status(task); - trace_rpc_bind_status(task); if (task->tk_status >= 0) goto out_next; if (xprt_bound(xprt)) { @@ -1994,12 +1966,10 @@ call_bind_status(struct rpc_task *task) switch (task->tk_status) { case -ENOMEM: - dprintk("RPC: %5u rpcbind out of memory\n", task->tk_pid); rpc_delay(task, HZ >> 2); goto retry_timeout; case -EACCES: - dprintk("RPC: %5u remote rpcbind: RPC program/version " - "unavailable\n", task->tk_pid); + trace_rpcb_prog_unavail_err(task); /* fail immediately if this is an RPC ping */ if (task->tk_msg.rpc_proc->p_proc == 0) { status = -EOPNOTSUPP; @@ -2016,17 +1986,14 @@ call_bind_status(struct rpc_task *task) case -EAGAIN: goto retry_timeout; case -ETIMEDOUT: - dprintk("RPC: %5u rpcbind request timed out\n", - task->tk_pid); + trace_rpcb_timeout_err(task); goto retry_timeout; case -EPFNOSUPPORT: /* server doesn't support any rpcbind version we know of */ - dprintk("RPC: %5u unrecognized remote rpcbind service\n", - task->tk_pid); + trace_rpcb_bind_version_err(task); break; case -EPROTONOSUPPORT: - dprintk("RPC: %5u remote rpcbind version unavailable, retrying\n", - task->tk_pid); + trace_rpcb_bind_version_err(task); goto retry_timeout; case -ECONNREFUSED: /* connection problems */ case -ECONNRESET: @@ -2037,8 +2004,7 @@ call_bind_status(struct rpc_task *task) case -EHOSTUNREACH: case -ENETUNREACH: case -EPIPE: - dprintk("RPC: %5u remote rpcbind unreachable: %d\n", - task->tk_pid, task->tk_status); + trace_rpcb_unreachable_err(task); if (!RPC_IS_SOFTCONN(task)) { rpc_delay(task, 5*HZ); goto retry_timeout; @@ -2046,8 +2012,7 @@ call_bind_status(struct rpc_task *task) status = task->tk_status; break; default: - dprintk("RPC: %5u unrecognized rpcbind error (%d)\n", - task->tk_pid, -task->tk_status); + trace_rpcb_unrecognized_err(task); } rpc_call_rpcerror(task, status); @@ -2079,10 +2044,6 @@ call_connect(struct rpc_task *task) return; } - dprintk("RPC: %5u call_connect xprt %p %s connected\n", - task->tk_pid, xprt, - (xprt_connected(xprt) ? "is" : "is not")); - task->tk_action = call_connect_status; if (task->tk_status < 0) return; @@ -2110,7 +2071,6 @@ call_connect_status(struct rpc_task *task) return; } - dprint_status(task); trace_rpc_connect_status(task); if (task->tk_status == 0) { @@ -2178,8 +2138,6 @@ call_transmit(struct rpc_task *task) return; } - dprint_status(task); - task->tk_action = call_transmit_status; if (!xprt_prepare_transmit(task)) return; @@ -2214,7 +2172,6 @@ call_transmit_status(struct rpc_task *task) switch (task->tk_status) { default: - dprint_status(task); break; case -EBADMSG: task->tk_status = 0; @@ -2296,8 +2253,6 @@ call_bc_transmit_status(struct rpc_task *task) if (rpc_task_transmitted(task)) task->tk_status = 0; - dprint_status(task); - switch (task->tk_status) { case 0: /* Success */ @@ -2357,8 +2312,6 @@ call_status(struct rpc_task *task) if (!task->tk_msg.rpc_proc->p_proc) trace_xprt_ping(task->tk_xprt, task->tk_status); - dprint_status(task); - status = task->tk_status; if (status >= 0) { task->tk_action = call_decode; @@ -2405,7 +2358,8 @@ call_status(struct rpc_task *task) goto out_exit; } task->tk_action = call_encode; - rpc_check_timeout(task); + if (status != -ECONNRESET && status != -ECONNABORTED) + rpc_check_timeout(task); return; out_exit: rpc_call_rpcerror(task, status); @@ -2433,7 +2387,7 @@ rpc_check_timeout(struct rpc_task *task) if (xprt_adjust_timeout(task->tk_rqstp) == 0) return; - dprintk("RPC: %5u call_timeout (major)\n", task->tk_pid); + trace_rpc_timeout_status(task); task->tk_timeouts++; if (RPC_IS_SOFTCONN(task) && !rpc_check_connected(task->tk_rqstp)) { @@ -2492,8 +2446,6 @@ call_decode(struct rpc_task *task) struct xdr_stream xdr; int err; - dprint_status(task); - if (!task->tk_msg.rpc_proc->p_decode) { task->tk_action = rpc_exit_task; return; @@ -2537,8 +2489,6 @@ out: case 0: task->tk_action = rpc_exit_task; task->tk_status = rpcauth_unwrap_resp(task, &xdr); - dprintk("RPC: %5u %s result %d\n", - task->tk_pid, __func__, task->tk_status); return; case -EAGAIN: task->tk_status = 0; diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c index 4a67685c83eb..38fe2ce8a5aa 100644 --- a/net/sunrpc/rpcb_clnt.c +++ b/net/sunrpc/rpcb_clnt.c @@ -31,11 +31,9 @@ #include <linux/sunrpc/sched.h> #include <linux/sunrpc/xprtsock.h> -#include "netns.h" +#include <trace/events/sunrpc.h> -#if IS_ENABLED(CONFIG_SUNRPC_DEBUG) -# define RPCDBG_FACILITY RPCDBG_BIND -#endif +#include "netns.h" #define RPCBIND_SOCK_PATHNAME "/var/run/rpcbind.sock" @@ -216,10 +214,6 @@ static void rpcb_set_local(struct net *net, struct rpc_clnt *clnt, sn->rpcb_is_af_local = is_af_local ? 1 : 0; smp_wmb(); sn->rpcb_users = 1; - dprintk("RPC: created new rpcb local clients (rpcb_local_clnt: " - "%p, rpcb_local_clnt4: %p) for net %x%s\n", - sn->rpcb_local_clnt, sn->rpcb_local_clnt4, - net->ns.inum, (net == &init_net) ? " (init_net)" : ""); } /* @@ -261,19 +255,13 @@ static int rpcb_create_local_unix(struct net *net) */ clnt = rpc_create(&args); if (IS_ERR(clnt)) { - dprintk("RPC: failed to create AF_LOCAL rpcbind " - "client (errno %ld).\n", PTR_ERR(clnt)); result = PTR_ERR(clnt); goto out; } clnt4 = rpc_bind_new_program(clnt, &rpcb_program, RPCBVERS_4); - if (IS_ERR(clnt4)) { - dprintk("RPC: failed to bind second program to " - "rpcbind v4 client (errno %ld).\n", - PTR_ERR(clnt4)); + if (IS_ERR(clnt4)) clnt4 = NULL; - } rpcb_set_local(net, clnt, clnt4, true); @@ -309,8 +297,6 @@ static int rpcb_create_local_net(struct net *net) clnt = rpc_create(&args); if (IS_ERR(clnt)) { - dprintk("RPC: failed to create local rpcbind " - "client (errno %ld).\n", PTR_ERR(clnt)); result = PTR_ERR(clnt); goto out; } @@ -321,12 +307,8 @@ static int rpcb_create_local_net(struct net *net) * v4 upcalls. */ clnt4 = rpc_bind_new_program(clnt, &rpcb_program, RPCBVERS_4); - if (IS_ERR(clnt4)) { - dprintk("RPC: failed to bind second program to " - "rpcbind v4 client (errno %ld).\n", - PTR_ERR(clnt4)); + if (IS_ERR(clnt4)) clnt4 = NULL; - } rpcb_set_local(net, clnt, clnt4, false); @@ -403,11 +385,8 @@ static int rpcb_register_call(struct sunrpc_net *sn, struct rpc_clnt *clnt, stru msg->rpc_resp = &result; error = rpc_call_sync(clnt, msg, flags); - if (error < 0) { - dprintk("RPC: failed to contact local rpcbind " - "server (errno %d).\n", -error); + if (error < 0) return error; - } if (!result) return -EACCES; @@ -461,9 +440,7 @@ int rpcb_register(struct net *net, u32 prog, u32 vers, int prot, unsigned short struct sunrpc_net *sn = net_generic(net, sunrpc_net_id); bool is_set = false; - dprintk("RPC: %sregistering (%u, %u, %d, %u) with local " - "rpcbind\n", (port ? "" : "un"), - prog, vers, prot, port); + trace_pmap_register(prog, vers, prot, port); msg.rpc_proc = &rpcb_procedures2[RPCBPROC_UNSET]; if (port != 0) { @@ -489,11 +466,6 @@ static int rpcb_register_inet4(struct sunrpc_net *sn, map->r_addr = rpc_sockaddr2uaddr(sap, GFP_KERNEL); - dprintk("RPC: %sregistering [%u, %u, %s, '%s'] with " - "local rpcbind\n", (port ? "" : "un"), - map->r_prog, map->r_vers, - map->r_addr, map->r_netid); - msg->rpc_proc = &rpcb_procedures4[RPCBPROC_UNSET]; if (port != 0) { msg->rpc_proc = &rpcb_procedures4[RPCBPROC_SET]; @@ -520,11 +492,6 @@ static int rpcb_register_inet6(struct sunrpc_net *sn, map->r_addr = rpc_sockaddr2uaddr(sap, GFP_KERNEL); - dprintk("RPC: %sregistering [%u, %u, %s, '%s'] with " - "local rpcbind\n", (port ? "" : "un"), - map->r_prog, map->r_vers, - map->r_addr, map->r_netid); - msg->rpc_proc = &rpcb_procedures4[RPCBPROC_UNSET]; if (port != 0) { msg->rpc_proc = &rpcb_procedures4[RPCBPROC_SET]; @@ -541,9 +508,7 @@ static int rpcb_unregister_all_protofamilies(struct sunrpc_net *sn, { struct rpcbind_args *map = msg->rpc_argp; - dprintk("RPC: unregistering [%u, %u, '%s'] with " - "local rpcbind\n", - map->r_prog, map->r_vers, map->r_netid); + trace_rpcb_unregister(map->r_prog, map->r_vers, map->r_netid); map->r_addr = ""; msg->rpc_proc = &rpcb_procedures4[RPCBPROC_UNSET]; @@ -615,6 +580,8 @@ int rpcb_v4_register(struct net *net, const u32 program, const u32 version, if (address == NULL) return rpcb_unregister_all_protofamilies(sn, &msg); + trace_rpcb_register(map.r_prog, map.r_vers, map.r_addr, map.r_netid); + switch (address->sa_family) { case AF_INET: return rpcb_register_inet4(sn, address, &msg); @@ -693,18 +660,12 @@ void rpcb_getport_async(struct rpc_task *task) rcu_read_unlock(); xprt = xprt_get(task->tk_xprt); - dprintk("RPC: %5u %s(%s, %u, %u, %d)\n", - task->tk_pid, __func__, - xprt->servername, clnt->cl_prog, clnt->cl_vers, xprt->prot); - /* Put self on the wait queue to ensure we get notified if * some other task is already attempting to bind the port */ rpc_sleep_on_timeout(&xprt->binding, task, NULL, jiffies + xprt->bind_timeout); if (xprt_test_and_set_binding(xprt)) { - dprintk("RPC: %5u %s: waiting for another binder\n", - task->tk_pid, __func__); xprt_put(xprt); return; } @@ -712,8 +673,6 @@ void rpcb_getport_async(struct rpc_task *task) /* Someone else may have bound if we slept */ if (xprt_bound(xprt)) { status = 0; - dprintk("RPC: %5u %s: already bound\n", - task->tk_pid, __func__); goto bailout_nofree; } @@ -732,20 +691,15 @@ void rpcb_getport_async(struct rpc_task *task) break; default: status = -EAFNOSUPPORT; - dprintk("RPC: %5u %s: bad address family\n", - task->tk_pid, __func__); goto bailout_nofree; } if (proc == NULL) { xprt->bind_index = 0; status = -EPFNOSUPPORT; - dprintk("RPC: %5u %s: no more getport versions available\n", - task->tk_pid, __func__); goto bailout_nofree; } - dprintk("RPC: %5u %s: trying rpcbind version %u\n", - task->tk_pid, __func__, bind_version); + trace_rpcb_getport(clnt, task, bind_version); rpcb_clnt = rpcb_create(xprt->xprt_net, clnt->cl_nodename, @@ -754,16 +708,12 @@ void rpcb_getport_async(struct rpc_task *task) clnt->cl_cred); if (IS_ERR(rpcb_clnt)) { status = PTR_ERR(rpcb_clnt); - dprintk("RPC: %5u %s: rpcb_create failed, error %ld\n", - task->tk_pid, __func__, PTR_ERR(rpcb_clnt)); goto bailout_nofree; } map = kzalloc(sizeof(struct rpcbind_args), GFP_NOFS); if (!map) { status = -ENOMEM; - dprintk("RPC: %5u %s: no memory available\n", - task->tk_pid, __func__); goto bailout_release_client; } map->r_prog = clnt->cl_prog; @@ -780,8 +730,6 @@ void rpcb_getport_async(struct rpc_task *task) map->r_addr = rpc_sockaddr2uaddr(sap, GFP_NOFS); if (!map->r_addr) { status = -ENOMEM; - dprintk("RPC: %5u %s: no memory available\n", - task->tk_pid, __func__); goto bailout_free_args; } map->r_owner = ""; @@ -818,34 +766,33 @@ static void rpcb_getport_done(struct rpc_task *child, void *data) { struct rpcbind_args *map = data; struct rpc_xprt *xprt = map->r_xprt; - int status = child->tk_status; + + map->r_status = child->tk_status; /* Garbage reply: retry with a lesser rpcbind version */ - if (status == -EIO) - status = -EPROTONOSUPPORT; + if (map->r_status == -EIO) + map->r_status = -EPROTONOSUPPORT; /* rpcbind server doesn't support this rpcbind protocol version */ - if (status == -EPROTONOSUPPORT) + if (map->r_status == -EPROTONOSUPPORT) xprt->bind_index++; - if (status < 0) { + if (map->r_status < 0) { /* rpcbind server not available on remote host? */ - xprt->ops->set_port(xprt, 0); + map->r_port = 0; + } else if (map->r_port == 0) { /* Requested RPC service wasn't registered on remote host */ - xprt->ops->set_port(xprt, 0); - status = -EACCES; + map->r_status = -EACCES; } else { /* Succeeded */ - xprt->ops->set_port(xprt, map->r_port); - xprt_set_bound(xprt); - status = 0; + map->r_status = 0; } - dprintk("RPC: %5u rpcb_getport_done(status %d, port %u)\n", - child->tk_pid, status, map->r_port); - - map->r_status = status; + trace_rpcb_setport(child, map->r_status, map->r_port); + xprt->ops->set_port(xprt, map->r_port); + if (map->r_port) + xprt_set_bound(xprt); } /* @@ -858,11 +805,6 @@ static void rpcb_enc_mapping(struct rpc_rqst *req, struct xdr_stream *xdr, const struct rpcbind_args *rpcb = data; __be32 *p; - dprintk("RPC: %5u encoding PMAP_%s call (%u, %u, %d, %u)\n", - req->rq_task->tk_pid, - req->rq_task->tk_msg.rpc_proc->p_name, - rpcb->r_prog, rpcb->r_vers, rpcb->r_prot, rpcb->r_port); - p = xdr_reserve_space(xdr, RPCB_mappingargs_sz << 2); *p++ = cpu_to_be32(rpcb->r_prog); *p++ = cpu_to_be32(rpcb->r_vers); @@ -884,8 +826,6 @@ static int rpcb_dec_getport(struct rpc_rqst *req, struct xdr_stream *xdr, return -EIO; port = be32_to_cpup(p); - dprintk("RPC: %5u PMAP_%s result: %lu\n", req->rq_task->tk_pid, - req->rq_task->tk_msg.rpc_proc->p_name, port); if (unlikely(port > USHRT_MAX)) return -EIO; @@ -906,11 +846,6 @@ static int rpcb_dec_set(struct rpc_rqst *req, struct xdr_stream *xdr, *boolp = 0; if (*p != xdr_zero) *boolp = 1; - - dprintk("RPC: %5u RPCB_%s call %s\n", - req->rq_task->tk_pid, - req->rq_task->tk_msg.rpc_proc->p_name, - (*boolp ? "succeeded" : "failed")); return 0; } @@ -935,12 +870,6 @@ static void rpcb_enc_getaddr(struct rpc_rqst *req, struct xdr_stream *xdr, const struct rpcbind_args *rpcb = data; __be32 *p; - dprintk("RPC: %5u encoding RPCB_%s call (%u, %u, '%s', '%s')\n", - req->rq_task->tk_pid, - req->rq_task->tk_msg.rpc_proc->p_name, - rpcb->r_prog, rpcb->r_vers, - rpcb->r_netid, rpcb->r_addr); - p = xdr_reserve_space(xdr, (RPCB_program_sz + RPCB_version_sz) << 2); *p++ = cpu_to_be32(rpcb->r_prog); *p = cpu_to_be32(rpcb->r_vers); @@ -970,11 +899,8 @@ static int rpcb_dec_getaddr(struct rpc_rqst *req, struct xdr_stream *xdr, * If the returned universal address is a null string, * the requested RPC service was not registered. */ - if (len == 0) { - dprintk("RPC: %5u RPCB reply: program not registered\n", - req->rq_task->tk_pid); + if (len == 0) return 0; - } if (unlikely(len > RPCBIND_MAXUADDRLEN)) goto out_fail; @@ -982,8 +908,6 @@ static int rpcb_dec_getaddr(struct rpc_rqst *req, struct xdr_stream *xdr, p = xdr_inline_decode(xdr, len); if (unlikely(p == NULL)) goto out_fail; - dprintk("RPC: %5u RPCB_%s reply: %*pE\n", req->rq_task->tk_pid, - req->rq_task->tk_msg.rpc_proc->p_name, len, (char *)p); if (rpc_uaddr2sockaddr(req->rq_xprt->xprt_net, (char *)p, len, sap, sizeof(address)) == 0) @@ -993,9 +917,6 @@ static int rpcb_dec_getaddr(struct rpc_rqst *req, struct xdr_stream *xdr, return 0; out_fail: - dprintk("RPC: %5u malformed RPCB_%s reply\n", - req->rq_task->tk_pid, - req->rq_task->tk_msg.rpc_proc->p_name); return -EIO; } diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index 7eba20a88438..f06d7c315017 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -27,10 +27,6 @@ #include "sunrpc.h" -#if IS_ENABLED(CONFIG_SUNRPC_DEBUG) -#define RPCDBG_FACILITY RPCDBG_SCHED -#endif - #define CREATE_TRACE_POINTS #include <trace/events/sunrpc.h> @@ -85,7 +81,6 @@ __rpc_disable_timer(struct rpc_wait_queue *queue, struct rpc_task *task) { if (list_empty(&task->u.tk_wait.timer_list)) return; - dprintk("RPC: %5u disabling timer\n", task->tk_pid); task->tk_timeout = 0; list_del(&task->u.tk_wait.timer_list); if (list_empty(&queue->timer_list.list)) @@ -111,9 +106,6 @@ static void __rpc_add_timer(struct rpc_wait_queue *queue, struct rpc_task *task, unsigned long timeout) { - dprintk("RPC: %5u setting alarm for %u ms\n", - task->tk_pid, jiffies_to_msecs(timeout - jiffies)); - task->tk_timeout = timeout; if (list_empty(&queue->timer_list.list) || time_before(timeout, queue->timer_list.expires)) rpc_set_queue_timer(queue, timeout); @@ -216,9 +208,6 @@ static void __rpc_add_wait_queue(struct rpc_wait_queue *queue, /* barrier matches the read in rpc_wake_up_task_queue_locked() */ smp_wmb(); rpc_set_queued(task); - - dprintk("RPC: %5u added to queue %p \"%s\"\n", - task->tk_pid, queue, rpc_qname(queue)); } /* @@ -241,8 +230,6 @@ static void __rpc_remove_wait_queue(struct rpc_wait_queue *queue, struct rpc_tas else list_del(&task->u.tk_wait.list); queue->qlen--; - dprintk("RPC: %5u removed from queue %p \"%s\"\n", - task->tk_pid, queue, rpc_qname(queue)); } static void __rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const char *qname, unsigned char nr_queues) @@ -382,13 +369,9 @@ static void __rpc_do_sleep_on_priority(struct rpc_wait_queue *q, struct rpc_task *task, unsigned char queue_priority) { - dprintk("RPC: %5u sleep_on(queue \"%s\" time %lu)\n", - task->tk_pid, rpc_qname(q), jiffies); - trace_rpc_task_sleep(task, q); __rpc_add_wait_queue(q, task, queue_priority); - } static void __rpc_sleep_on_priority(struct rpc_wait_queue *q, @@ -510,9 +493,6 @@ static void __rpc_do_wake_up_task_on_wq(struct workqueue_struct *wq, struct rpc_wait_queue *queue, struct rpc_task *task) { - dprintk("RPC: %5u __rpc_wake_up_task (now %lu)\n", - task->tk_pid, jiffies); - /* Has the task been executed yet? If not, we cannot wake it up! */ if (!RPC_IS_ACTIVATED(task)) { printk(KERN_ERR "RPC: Inactive task (%p) being woken up!\n", task); @@ -524,8 +504,6 @@ static void __rpc_do_wake_up_task_on_wq(struct workqueue_struct *wq, __rpc_remove_wait_queue(queue, task); rpc_make_runnable(wq, task); - - dprintk("RPC: __rpc_wake_up_task done\n"); } /* @@ -663,8 +641,6 @@ struct rpc_task *rpc_wake_up_first_on_wq(struct workqueue_struct *wq, { struct rpc_task *task = NULL; - dprintk("RPC: wake_up_first(%p \"%s\")\n", - queue, rpc_qname(queue)); spin_lock(&queue->lock); task = __rpc_find_next_queued(queue); if (task != NULL) @@ -770,7 +746,7 @@ static void __rpc_queue_timer_fn(struct work_struct *work) list_for_each_entry_safe(task, n, &queue->timer_list.list, u.tk_wait.timer_list) { timeo = task->tk_timeout; if (time_after_eq(now, timeo)) { - dprintk("RPC: %5u timeout\n", task->tk_pid); + trace_rpc_task_timeout(task, task->tk_action); task->tk_status = -ETIMEDOUT; rpc_wake_up_task_queue_locked(queue, task); continue; @@ -885,9 +861,6 @@ static void __rpc_execute(struct rpc_task *task) int task_is_async = RPC_IS_ASYNC(task); int status = 0; - dprintk("RPC: %5u __rpc_execute flags=0x%x\n", - task->tk_pid, task->tk_flags); - WARN_ON_ONCE(RPC_IS_QUEUED(task)); if (RPC_IS_QUEUED(task)) return; @@ -947,7 +920,7 @@ static void __rpc_execute(struct rpc_task *task) return; /* sync task: sleep here */ - dprintk("RPC: %5u sync task going to sleep\n", task->tk_pid); + trace_rpc_task_sync_sleep(task, task->tk_action); status = out_of_line_wait_on_bit(&task->tk_runstate, RPC_TASK_QUEUED, rpc_wait_bit_killable, TASK_KILLABLE); @@ -963,11 +936,9 @@ static void __rpc_execute(struct rpc_task *task) task->tk_rpc_status = -ERESTARTSYS; rpc_exit(task, -ERESTARTSYS); } - dprintk("RPC: %5u sync task resuming\n", task->tk_pid); + trace_rpc_task_sync_wake(task, task->tk_action); } - dprintk("RPC: %5u return %d, status %d\n", task->tk_pid, status, - task->tk_status); /* Release all resources associated with the task */ rpc_release_task(task); } @@ -1036,8 +1007,6 @@ int rpc_malloc(struct rpc_task *task) return -ENOMEM; buf->len = size; - dprintk("RPC: %5u allocated buffer of size %zu at %p\n", - task->tk_pid, size, buf); rqst->rq_buffer = buf->data; rqst->rq_rbuffer = (char *)rqst->rq_buffer + rqst->rq_callsize; return 0; @@ -1058,9 +1027,6 @@ void rpc_free(struct rpc_task *task) buf = container_of(buffer, struct rpc_buffer, data); size = buf->len; - dprintk("RPC: freeing buffer of size %zu at %p\n", - size, buf); - if (size <= RPC_BUFFER_MAXSIZE) mempool_free(buf, rpc_buffer_mempool); else @@ -1095,9 +1061,6 @@ static void rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *ta task->tk_action = rpc_prepare_task; rpc_init_task_statistics(task); - - dprintk("RPC: new task initialized, procpid %u\n", - task_pid_nr(current)); } static struct rpc_task * @@ -1121,7 +1084,6 @@ struct rpc_task *rpc_new_task(const struct rpc_task_setup *setup_data) rpc_init_task(task, setup_data); task->tk_flags |= flags; - dprintk("RPC: allocated task %p\n", task); return task; } @@ -1151,10 +1113,8 @@ static void rpc_free_task(struct rpc_task *task) put_rpccred(task->tk_op_cred); rpc_release_calldata(task->tk_ops, task->tk_calldata); - if (tk_flags & RPC_TASK_DYNAMIC) { - dprintk("RPC: %5u freeing task\n", task->tk_pid); + if (tk_flags & RPC_TASK_DYNAMIC) mempool_free(task, rpc_task_mempool); - } } static void rpc_async_release(struct work_struct *work) @@ -1208,8 +1168,6 @@ EXPORT_SYMBOL_GPL(rpc_put_task_async); static void rpc_release_task(struct rpc_task *task) { - dprintk("RPC: %5u release task\n", task->tk_pid); - WARN_ON_ONCE(RPC_IS_QUEUED(task)); rpc_release_resources_task(task); @@ -1250,7 +1208,6 @@ static int rpciod_start(void) /* * Create the rpciod thread and wait for it to start. */ - dprintk("RPC: creating workqueue rpciod\n"); wq = alloc_workqueue("rpciod", WQ_MEM_RECLAIM | WQ_UNBOUND, 0); if (!wq) goto out_failed; @@ -1275,7 +1232,6 @@ static void rpciod_stop(void) if (rpciod_workqueue == NULL) return; - dprintk("RPC: destroying workqueue rpciod\n"); wq = rpciod_workqueue; rpciod_workqueue = NULL; diff --git a/net/sunrpc/sunrpc.h b/net/sunrpc/sunrpc.h index f6fe2e6cd65a..2f59464e6524 100644 --- a/net/sunrpc/sunrpc.h +++ b/net/sunrpc/sunrpc.h @@ -4,7 +4,7 @@ NetApp provides this source code under the GPL v2 License. The GPL v2 license is available at -http://opensource.org/licenses/gpl-license.php. +https://opensource.org/licenses/gpl-license.php. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c index be11d672b5b9..3feff529a764 100644 --- a/net/sunrpc/xdr.c +++ b/net/sunrpc/xdr.c @@ -19,6 +19,9 @@ #include <linux/bvec.h> #include <trace/events/sunrpc.h> +static void _copy_to_pages(struct page **, size_t, const char *, size_t); + + /* * XDR functions for basic NFS types */ @@ -202,6 +205,88 @@ EXPORT_SYMBOL_GPL(xdr_inline_pages); */ /** + * _shift_data_left_pages + * @pages: vector of pages containing both the source and dest memory area. + * @pgto_base: page vector address of destination + * @pgfrom_base: page vector address of source + * @len: number of bytes to copy + * + * Note: the addresses pgto_base and pgfrom_base are both calculated in + * the same way: + * if a memory area starts at byte 'base' in page 'pages[i]', + * then its address is given as (i << PAGE_CACHE_SHIFT) + base + * Alse note: pgto_base must be < pgfrom_base, but the memory areas + * they point to may overlap. + */ +static void +_shift_data_left_pages(struct page **pages, size_t pgto_base, + size_t pgfrom_base, size_t len) +{ + struct page **pgfrom, **pgto; + char *vfrom, *vto; + size_t copy; + + BUG_ON(pgfrom_base <= pgto_base); + + pgto = pages + (pgto_base >> PAGE_SHIFT); + pgfrom = pages + (pgfrom_base >> PAGE_SHIFT); + + pgto_base &= ~PAGE_MASK; + pgfrom_base &= ~PAGE_MASK; + + do { + if (pgto_base >= PAGE_SIZE) { + pgto_base = 0; + pgto++; + } + if (pgfrom_base >= PAGE_SIZE){ + pgfrom_base = 0; + pgfrom++; + } + + copy = len; + if (copy > (PAGE_SIZE - pgto_base)) + copy = PAGE_SIZE - pgto_base; + if (copy > (PAGE_SIZE - pgfrom_base)) + copy = PAGE_SIZE - pgfrom_base; + + vto = kmap_atomic(*pgto); + if (*pgto != *pgfrom) { + vfrom = kmap_atomic(*pgfrom); + memcpy(vto + pgto_base, vfrom + pgfrom_base, copy); + kunmap_atomic(vfrom); + } else + memmove(vto + pgto_base, vto + pgfrom_base, copy); + flush_dcache_page(*pgto); + kunmap_atomic(vto); + + pgto_base += copy; + pgfrom_base += copy; + + } while ((len -= copy) != 0); +} + +static void +_shift_data_left_tail(struct xdr_buf *buf, unsigned int pgto, size_t len) +{ + struct kvec *tail = buf->tail; + + if (len > tail->iov_len) + len = tail->iov_len; + + _copy_to_pages(buf->pages, + buf->page_base + pgto, + (char *)tail->iov_base, + len); + tail->iov_len -= len; + + if (tail->iov_len > 0) + memmove((char *)tail->iov_base, + tail->iov_base + len, + tail->iov_len); +} + +/** * _shift_data_right_pages * @pages: vector of pages containing both the source and dest memory area. * @pgto_base: page vector address of destination @@ -266,6 +351,46 @@ _shift_data_right_pages(struct page **pages, size_t pgto_base, } while ((len -= copy) != 0); } +static unsigned int +_shift_data_right_tail(struct xdr_buf *buf, unsigned int pgfrom, size_t len) +{ + struct kvec *tail = buf->tail; + unsigned int tailbuf_len; + unsigned int result = 0; + size_t copy; + + tailbuf_len = buf->buflen - buf->head->iov_len - buf->page_len; + + /* Shift the tail first */ + if (tailbuf_len != 0) { + unsigned int free_space = tailbuf_len - tail->iov_len; + + if (len < free_space) + free_space = len; + if (len > free_space) + len = free_space; + + tail->iov_len += free_space; + copy = len; + + if (tail->iov_len > len) { + char *p = (char *)tail->iov_base + len; + memmove(p, tail->iov_base, tail->iov_len - free_space); + result += tail->iov_len - free_space; + } else + copy = tail->iov_len; + + /* Copy from the inlined pages into the tail */ + _copy_from_pages((char *)tail->iov_base, + buf->pages, + buf->page_base + pgfrom, + copy); + result += copy; + } + + return result; +} + /** * _copy_to_pages * @pages: array of pages @@ -351,6 +476,38 @@ _copy_from_pages(char *p, struct page **pages, size_t pgbase, size_t len) EXPORT_SYMBOL_GPL(_copy_from_pages); /** + * _zero_pages + * @pages: array of pages + * @pgbase: beginning page vector address + * @len: length + */ +static void +_zero_pages(struct page **pages, size_t pgbase, size_t len) +{ + struct page **page; + char *vpage; + size_t zero; + + page = pages + (pgbase >> PAGE_SHIFT); + pgbase &= ~PAGE_MASK; + + do { + zero = PAGE_SIZE - pgbase; + if (zero > len) + zero = len; + + vpage = kmap_atomic(*page); + memset(vpage + pgbase, 0, zero); + kunmap_atomic(vpage); + + flush_dcache_page(*page); + pgbase = 0; + page++; + + } while ((len -= zero) != 0); +} + +/** * xdr_shrink_bufhead * @buf: xdr_buf * @len: bytes to remove from buf->head[0] @@ -446,39 +603,13 @@ xdr_shrink_bufhead(struct xdr_buf *buf, size_t len) static unsigned int xdr_shrink_pagelen(struct xdr_buf *buf, size_t len) { - struct kvec *tail; - size_t copy; unsigned int pglen = buf->page_len; - unsigned int tailbuf_len; unsigned int result; - result = 0; - tail = buf->tail; if (len > buf->page_len) len = buf-> page_len; - tailbuf_len = buf->buflen - buf->head->iov_len - buf->page_len; - - /* Shift the tail first */ - if (tailbuf_len != 0) { - unsigned int free_space = tailbuf_len - tail->iov_len; - if (len < free_space) - free_space = len; - tail->iov_len += free_space; - - copy = len; - if (tail->iov_len > len) { - char *p = (char *)tail->iov_base + len; - memmove(p, tail->iov_base, tail->iov_len - len); - result += tail->iov_len - len; - } else - copy = tail->iov_len; - /* Copy from the inlined pages into the tail */ - _copy_from_pages((char *)tail->iov_base, - buf->pages, buf->page_base + pglen - len, - copy); - result += copy; - } + result = _shift_data_right_tail(buf, pglen - len, len); buf->page_len -= len; buf->buflen -= len; /* Have we truncated the message? */ @@ -506,6 +637,19 @@ unsigned int xdr_stream_pos(const struct xdr_stream *xdr) EXPORT_SYMBOL_GPL(xdr_stream_pos); /** + * xdr_page_pos - Return the current offset from the start of the xdr pages + * @xdr: pointer to struct xdr_stream + */ +unsigned int xdr_page_pos(const struct xdr_stream *xdr) +{ + unsigned int pos = xdr_stream_pos(xdr); + + WARN_ON(pos < xdr->buf->head[0].iov_len); + return pos - xdr->buf->head[0].iov_len; +} +EXPORT_SYMBOL_GPL(xdr_page_pos); + +/** * xdr_init_encode - Initialize a struct xdr_stream for sending data. * @xdr: pointer to xdr_stream struct * @buf: pointer to XDR buffer in which to encode data @@ -825,6 +969,13 @@ static int xdr_set_page_base(struct xdr_stream *xdr, return 0; } +static void xdr_set_page(struct xdr_stream *xdr, unsigned int base, + unsigned int len) +{ + if (xdr_set_page_base(xdr, base, len) < 0) + xdr_set_iov(xdr, xdr->buf->tail, xdr->nwords << 2); +} + static void xdr_set_next_page(struct xdr_stream *xdr) { unsigned int newbase; @@ -832,8 +983,7 @@ static void xdr_set_next_page(struct xdr_stream *xdr) newbase = (1 + xdr->page_ptr - xdr->buf->pages) << PAGE_SHIFT; newbase -= xdr->buf->page_base; - if (xdr_set_page_base(xdr, newbase, PAGE_SIZE) < 0) - xdr_set_iov(xdr, xdr->buf->tail, xdr->nwords << 2); + xdr_set_page(xdr, newbase, PAGE_SIZE); } static bool xdr_set_next_buffer(struct xdr_stream *xdr) @@ -841,8 +991,7 @@ static bool xdr_set_next_buffer(struct xdr_stream *xdr) if (xdr->page_ptr != NULL) xdr_set_next_page(xdr); else if (xdr->iov == xdr->buf->head) { - if (xdr_set_page_base(xdr, 0, PAGE_SIZE) < 0) - xdr_set_iov(xdr, xdr->buf->tail, xdr->nwords << 2); + xdr_set_page(xdr, 0, PAGE_SIZE); } return xdr->p != xdr->end; } @@ -979,26 +1128,33 @@ out_overflow: } EXPORT_SYMBOL_GPL(xdr_inline_decode); -static unsigned int xdr_align_pages(struct xdr_stream *xdr, unsigned int len) +static void xdr_realign_pages(struct xdr_stream *xdr) { struct xdr_buf *buf = xdr->buf; - struct kvec *iov; - unsigned int nwords = XDR_QUADLEN(len); + struct kvec *iov = buf->head; unsigned int cur = xdr_stream_pos(xdr); unsigned int copied, offset; - if (xdr->nwords == 0) - return 0; - /* Realign pages to current pointer position */ - iov = buf->head; if (iov->iov_len > cur) { offset = iov->iov_len - cur; copied = xdr_shrink_bufhead(buf, offset); trace_rpc_xdr_alignment(xdr, offset, copied); xdr->nwords = XDR_QUADLEN(buf->len - cur); } +} +static unsigned int xdr_align_pages(struct xdr_stream *xdr, unsigned int len) +{ + struct xdr_buf *buf = xdr->buf; + unsigned int nwords = XDR_QUADLEN(len); + unsigned int cur = xdr_stream_pos(xdr); + unsigned int copied, offset; + + if (xdr->nwords == 0) + return 0; + + xdr_realign_pages(xdr); if (nwords > xdr->nwords) { nwords = xdr->nwords; len = nwords << 2; @@ -1057,6 +1213,79 @@ unsigned int xdr_read_pages(struct xdr_stream *xdr, unsigned int len) } EXPORT_SYMBOL_GPL(xdr_read_pages); +uint64_t xdr_align_data(struct xdr_stream *xdr, uint64_t offset, uint32_t length) +{ + struct xdr_buf *buf = xdr->buf; + unsigned int from, bytes; + unsigned int shift = 0; + + if ((offset + length) < offset || + (offset + length) > buf->page_len) + length = buf->page_len - offset; + + xdr_realign_pages(xdr); + from = xdr_page_pos(xdr); + bytes = xdr->nwords << 2; + if (length < bytes) + bytes = length; + + /* Move page data to the left */ + if (from > offset) { + shift = min_t(unsigned int, bytes, buf->page_len - from); + _shift_data_left_pages(buf->pages, + buf->page_base + offset, + buf->page_base + from, + shift); + bytes -= shift; + + /* Move tail data into the pages, if necessary */ + if (bytes > 0) + _shift_data_left_tail(buf, offset + shift, bytes); + } + + xdr->nwords -= XDR_QUADLEN(length); + xdr_set_page(xdr, from + length, PAGE_SIZE); + return length; +} +EXPORT_SYMBOL_GPL(xdr_align_data); + +uint64_t xdr_expand_hole(struct xdr_stream *xdr, uint64_t offset, uint64_t length) +{ + struct xdr_buf *buf = xdr->buf; + unsigned int bytes; + unsigned int from; + unsigned int truncated = 0; + + if ((offset + length) < offset || + (offset + length) > buf->page_len) + length = buf->page_len - offset; + + xdr_realign_pages(xdr); + from = xdr_page_pos(xdr); + bytes = xdr->nwords << 2; + + if (offset + length + bytes > buf->page_len) { + unsigned int shift = (offset + length + bytes) - buf->page_len; + unsigned int res = _shift_data_right_tail(buf, from + bytes - shift, shift); + truncated = shift - res; + xdr->nwords -= XDR_QUADLEN(truncated); + bytes -= shift; + } + + /* Now move the page data over and zero pages */ + if (bytes > 0) + _shift_data_right_pages(buf->pages, + buf->page_base + offset + length, + buf->page_base + from, + bytes); + _zero_pages(buf->pages, buf->page_base + offset, length); + + buf->len += length - (from - offset) - truncated; + xdr_set_page(xdr, offset + length, PAGE_SIZE); + return length; +} +EXPORT_SYMBOL_GPL(xdr_expand_hole); + /** * xdr_enter_page - decode data from the XDR page * @xdr: pointer to xdr_stream struct diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index 5a8e47bbfb9f..f6c17e75f20e 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -834,8 +834,7 @@ void xprt_connect(struct rpc_task *task) { struct rpc_xprt *xprt = task->tk_rqstp->rq_xprt; - dprintk("RPC: %5u xprt_connect xprt %p %s connected\n", task->tk_pid, - xprt, (xprt_connected(xprt) ? "is" : "is not")); + trace_xprt_connect(xprt); if (!xprt_bound(xprt)) { task->tk_status = -EAGAIN; @@ -1131,8 +1130,6 @@ void xprt_complete_rqst(struct rpc_task *task, int copied) struct rpc_rqst *req = task->tk_rqstp; struct rpc_xprt *xprt = req->rq_xprt; - trace_xprt_complete_rqst(xprt, req->rq_xid, copied); - xprt->stat.recvs++; req->rq_private_buf.len = copied; @@ -1269,7 +1266,6 @@ xprt_request_enqueue_transmit(struct rpc_task *task) /* Note: req is added _before_ pos */ list_add_tail(&req->rq_xmit, &pos->rq_xmit); INIT_LIST_HEAD(&req->rq_xmit2); - trace_xprt_enq_xmit(task, 1); goto out; } } else if (RPC_IS_SWAPPER(task)) { @@ -1281,7 +1277,6 @@ xprt_request_enqueue_transmit(struct rpc_task *task) /* Note: req is added _before_ pos */ list_add_tail(&req->rq_xmit, &pos->rq_xmit); INIT_LIST_HEAD(&req->rq_xmit2); - trace_xprt_enq_xmit(task, 2); goto out; } } else if (!req->rq_seqno) { @@ -1290,13 +1285,11 @@ xprt_request_enqueue_transmit(struct rpc_task *task) continue; list_add_tail(&req->rq_xmit2, &pos->rq_xmit2); INIT_LIST_HEAD(&req->rq_xmit); - trace_xprt_enq_xmit(task, 3); goto out; } } list_add_tail(&req->rq_xmit, &xprt->xmit_queue); INIT_LIST_HEAD(&req->rq_xmit2); - trace_xprt_enq_xmit(task, 4); out: set_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate); spin_unlock(&xprt->queue_lock); @@ -1414,9 +1407,9 @@ bool xprt_prepare_transmit(struct rpc_task *task) struct rpc_rqst *req = task->tk_rqstp; struct rpc_xprt *xprt = req->rq_xprt; - dprintk("RPC: %5u xprt_prepare_transmit\n", task->tk_pid); - if (!xprt_lock_write(xprt, task)) { + trace_xprt_transmit_queued(xprt, task); + /* Race breaker: someone may have transmitted us */ if (!test_bit(RPC_TASK_NEED_XMIT, &task->tk_runstate)) rpc_wake_up_queued_task_set_status(&xprt->sending, @@ -1520,10 +1513,13 @@ xprt_transmit(struct rpc_task *task) { struct rpc_rqst *next, *req = task->tk_rqstp; struct rpc_xprt *xprt = req->rq_xprt; - int status; + int counter, status; spin_lock(&xprt->queue_lock); + counter = 0; while (!list_empty(&xprt->xmit_queue)) { + if (++counter == 20) + break; next = list_first_entry(&xprt->xmit_queue, struct rpc_rqst, rq_xmit); xprt_pin_rqst(next); @@ -1531,7 +1527,6 @@ xprt_transmit(struct rpc_task *task) status = xprt_request_transmit(next, task); if (status == -EBADMSG && next != req) status = 0; - cond_resched(); spin_lock(&xprt->queue_lock); xprt_unpin_rqst(next); if (status == 0) { @@ -1747,8 +1742,8 @@ xprt_request_init(struct rpc_task *task) req->rq_rcv_buf.bvec = NULL; req->rq_release_snd_buf = NULL; xprt_init_majortimeo(task, req); - dprintk("RPC: %5u reserved req %p xid %08x\n", task->tk_pid, - req, ntohl(req->rq_xid)); + + trace_xprt_reserve(req); } static void @@ -1838,7 +1833,6 @@ void xprt_release(struct rpc_task *task) if (req->rq_release_snd_buf) req->rq_release_snd_buf(req); - dprintk("RPC: %5u release request %p\n", task->tk_pid, req); if (likely(!bc_prealloc(req))) xprt->ops->free_slot(xprt, req); else diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c index 7f94c9a19fd3..44888f5badef 100644 --- a/net/sunrpc/xprtrdma/frwr_ops.c +++ b/net/sunrpc/xprtrdma/frwr_ops.c @@ -124,7 +124,7 @@ int frwr_mr_init(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr *mr) if (IS_ERR(frmr)) goto out_mr_err; - sg = kcalloc(depth, sizeof(*sg), GFP_NOFS); + sg = kmalloc_array(depth, sizeof(*sg), GFP_NOFS); if (!sg) goto out_list_err; diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c index 053c8ab1265a..8915e42240d3 100644 --- a/net/sunrpc/xprtrdma/transport.c +++ b/net/sunrpc/xprtrdma/transport.c @@ -413,9 +413,6 @@ xprt_rdma_set_port(struct rpc_xprt *xprt, u16 port) kfree(xprt->address_strings[RPC_DISPLAY_HEX_PORT]); snprintf(buf, sizeof(buf), "%4hx", port); xprt->address_strings[RPC_DISPLAY_HEX_PORT] = kstrdup(buf, GFP_KERNEL); - - trace_xprtrdma_op_setport(container_of(xprt, struct rpcrdma_xprt, - rx_xprt)); } /** @@ -586,11 +583,9 @@ xprt_rdma_allocate(struct rpc_task *task) rqst->rq_buffer = rdmab_data(req->rl_sendbuf); rqst->rq_rbuffer = rdmab_data(req->rl_recvbuf); - trace_xprtrdma_op_allocate(task, req); return 0; out_fail: - trace_xprtrdma_op_allocate(task, NULL); return -ENOMEM; } @@ -607,8 +602,6 @@ xprt_rdma_free(struct rpc_task *task) struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(rqst->rq_xprt); struct rpcrdma_req *req = rpcr_to_rdmar(rqst); - trace_xprtrdma_op_free(task, req); - if (!list_empty(&req->rl_registered)) frwr_unmap_sync(r_xprt, req); diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 554e1bb4c1c7..7090bbee0ec5 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -762,10 +762,7 @@ static int xs_nospace(struct rpc_rqst *req) struct sock *sk = transport->inet; int ret = -EAGAIN; - dprintk("RPC: %5u xmit incomplete (%u left of %u)\n", - req->rq_task->tk_pid, - req->rq_slen - transport->xmit.offset, - req->rq_slen); + trace_rpc_socket_nospace(req, transport); /* Protect against races with write_space */ spin_lock(&xprt->transport_lock); |