summaryrefslogtreecommitdiffstats
path: root/fs
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2013-11-16 21:04:02 +0100
committerLinus Torvalds <torvalds@linux-foundation.org>2013-11-16 21:04:02 +0100
commit449bf8d03c5b94f00cc014ff601c2fe2eebb5a6e (patch)
tree8f17959c0a20d9ca3061d28036855813c775c783 /fs
parentMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mas... (diff)
parentnfsd: export proper maximum file size to the client (diff)
downloadlinux-449bf8d03c5b94f00cc014ff601c2fe2eebb5a6e.tar.xz
linux-449bf8d03c5b94f00cc014ff601c2fe2eebb5a6e.zip
Merge branch 'nfsd-next' of git://linux-nfs.org/~bfields/linux
Pull nfsd changes from Bruce Fields: "This includes miscellaneous bugfixes and cleanup and a performance fix for write-heavy NFSv4 workloads. (The most significant nfsd-relevant change this time is actually in the delegation patches that went through Viro, fixing a long-standing bug that can cause NFSv4 clients to miss updates made by non-nfs users of the filesystem. Those enable some followup nfsd patches which I have queued locally, but those can wait till 3.14)" * 'nfsd-next' of git://linux-nfs.org/~bfields/linux: (24 commits) nfsd: export proper maximum file size to the client nfsd4: improve write performance with better sendspace reservations svcrpc: remove an unnecessary assignment sunrpc: comment typo fix Revert "nfsd: remove_stid can be incorporated into nfs4_put_delegation" nfsd4: fix discarded security labels on setattr NFSD: Add support for NFS v4.2 operation checking nfsd4: nfsd_shutdown_net needs state lock NFSD: Combine decode operations for v4 and v4.1 nfsd: -EINVAL on invalid anonuid/gid instead of silent failure nfsd: return better errors to exportfs nfsd: fh_update should error out in unexpected cases nfsd4: need to destroy revoked delegations in destroy_client nfsd: no need to unhash_stid before free nfsd: remove_stid can be incorporated into nfs4_put_delegation nfsd: nfs4_open_delegation needs to remove_stid rather than unhash_stid nfsd: nfs4_free_stid nfsd: fix Kconfig syntax sunrpc: trim off EC bytes in GSSAPI v2 unwrap gss_krb5: document that we ignore sequence number ...
Diffstat (limited to 'fs')
-rw-r--r--fs/nfsd/Kconfig2
-rw-r--r--fs/nfsd/export.c24
-rw-r--r--fs/nfsd/nfs4state.c26
-rw-r--r--fs/nfsd/nfs4xdr.c132
-rw-r--r--fs/nfsd/nfsfh.c8
5 files changed, 115 insertions, 77 deletions
diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig
index dc8f1ef665ce..f994e750e0d1 100644
--- a/fs/nfsd/Kconfig
+++ b/fs/nfsd/Kconfig
@@ -95,7 +95,7 @@ config NFSD_V4_SECURITY_LABEL
Smack policies on NFSv4 files, say N.
WARNING: there is still a chance of backwards-incompatible protocol changes.
- For now we recommend "Y" only for developers and testers."
+ For now we recommend "Y" only for developers and testers.
config NFSD_FAULT_INJECTION
bool "NFS server manual fault injection"
diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c
index 5f38ea36e266..8513c598fabf 100644
--- a/fs/nfsd/export.c
+++ b/fs/nfsd/export.c
@@ -536,16 +536,12 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen)
if (err)
goto out3;
exp.ex_anon_uid= make_kuid(&init_user_ns, an_int);
- if (!uid_valid(exp.ex_anon_uid))
- goto out3;
/* anon gid */
err = get_int(&mesg, &an_int);
if (err)
goto out3;
exp.ex_anon_gid= make_kgid(&init_user_ns, an_int);
- if (!gid_valid(exp.ex_anon_gid))
- goto out3;
/* fsid */
err = get_int(&mesg, &an_int);
@@ -583,6 +579,26 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen)
exp.ex_uuid);
if (err)
goto out4;
+ /*
+ * No point caching this if it would immediately expire.
+ * Also, this protects exportfs's dummy export from the
+ * anon_uid/anon_gid checks:
+ */
+ if (exp.h.expiry_time < seconds_since_boot())
+ goto out4;
+ /*
+ * For some reason exportfs has been passing down an
+ * invalid (-1) uid & gid on the "dummy" export which it
+ * uses to test export support. To make sure exportfs
+ * sees errors from check_export we therefore need to
+ * delay these checks till after check_export:
+ */
+ err = -EINVAL;
+ if (!uid_valid(exp.ex_anon_uid))
+ goto out4;
+ if (!gid_valid(exp.ex_anon_gid))
+ goto out4;
+ err = 0;
}
expp = svc_export_lookup(&exp);
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index f36a30a9f2d1..105d6fa7c514 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -402,11 +402,16 @@ static void remove_stid(struct nfs4_stid *s)
idr_remove(stateids, s->sc_stateid.si_opaque.so_id);
}
+static void nfs4_free_stid(struct kmem_cache *slab, struct nfs4_stid *s)
+{
+ kmem_cache_free(slab, s);
+}
+
void
nfs4_put_delegation(struct nfs4_delegation *dp)
{
if (atomic_dec_and_test(&dp->dl_count)) {
- kmem_cache_free(deleg_slab, dp);
+ nfs4_free_stid(deleg_slab, &dp->dl_stid);
num_delegations--;
}
}
@@ -610,7 +615,7 @@ static void close_generic_stateid(struct nfs4_ol_stateid *stp)
static void free_generic_stateid(struct nfs4_ol_stateid *stp)
{
remove_stid(&stp->st_stid);
- kmem_cache_free(stateid_slab, stp);
+ nfs4_free_stid(stateid_slab, &stp->st_stid);
}
static void release_lock_stateid(struct nfs4_ol_stateid *stp)
@@ -668,7 +673,6 @@ static void unhash_open_stateid(struct nfs4_ol_stateid *stp)
static void release_open_stateid(struct nfs4_ol_stateid *stp)
{
unhash_open_stateid(stp);
- unhash_stid(&stp->st_stid);
free_generic_stateid(stp);
}
@@ -690,7 +694,6 @@ static void release_last_closed_stateid(struct nfs4_openowner *oo)
struct nfs4_ol_stateid *s = oo->oo_last_closed_stid;
if (s) {
- unhash_stid(&s->st_stid);
free_generic_stateid(s);
oo->oo_last_closed_stid = NULL;
}
@@ -1127,6 +1130,11 @@ destroy_client(struct nfs4_client *clp)
dp = list_entry(reaplist.next, struct nfs4_delegation, dl_recall_lru);
destroy_delegation(dp);
}
+ list_splice_init(&clp->cl_revoked, &reaplist);
+ while (!list_empty(&reaplist)) {
+ dp = list_entry(reaplist.next, struct nfs4_delegation, dl_recall_lru);
+ destroy_revoked_delegation(dp);
+ }
while (!list_empty(&clp->cl_openowners)) {
oo = list_entry(clp->cl_openowners.next, struct nfs4_openowner, oo_perclient);
release_openowner(oo);
@@ -3154,7 +3162,7 @@ nfs4_open_delegation(struct net *net, struct svc_fh *fh,
open->op_delegate_type = NFS4_OPEN_DELEGATE_READ;
return;
out_free:
- unhash_stid(&dp->dl_stid);
+ remove_stid(&dp->dl_stid);
nfs4_put_delegation(dp);
out_no_deleg:
open->op_delegate_type = NFS4_OPEN_DELEGATE_NONE;
@@ -3995,10 +4003,9 @@ nfsd4_close(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
nfsd4_close_open_stateid(stp);
- if (cstate->minorversion) {
- unhash_stid(&stp->st_stid);
+ if (cstate->minorversion)
free_generic_stateid(stp);
- } else
+ else
oo->oo_last_closed_stid = stp;
if (list_empty(&oo->oo_owner.so_stateids)) {
@@ -5119,7 +5126,6 @@ out_recovery:
return ret;
}
-/* should be called with the state lock held */
void
nfs4_state_shutdown_net(struct net *net)
{
@@ -5130,6 +5136,7 @@ nfs4_state_shutdown_net(struct net *net)
cancel_delayed_work_sync(&nn->laundromat_work);
locks_end_grace(&nn->nfsd4_manager);
+ nfs4_lock_state();
INIT_LIST_HEAD(&reaplist);
spin_lock(&recall_lock);
list_for_each_safe(pos, next, &nn->del_recall_lru) {
@@ -5144,6 +5151,7 @@ nfs4_state_shutdown_net(struct net *net)
nfsd4_client_tracking_exit(net);
nfs4_state_destroy_net(net);
+ nfs4_unlock_state();
}
void
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index d9454fe5653f..088de1355e93 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -411,6 +411,7 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval,
label->data = kzalloc(dummy32 + 1, GFP_KERNEL);
if (!label->data)
return nfserr_jukebox;
+ label->len = dummy32;
defer_free(argp, kfree, label->data);
memcpy(label->data, buf, dummy32);
}
@@ -945,13 +946,16 @@ static __be32
nfsd4_decode_open_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_open_confirm *open_conf)
{
DECODE_HEAD;
-
+
+ if (argp->minorversion >= 1)
+ return nfserr_notsupp;
+
status = nfsd4_decode_stateid(argp, &open_conf->oc_req_stateid);
if (status)
return status;
READ_BUF(4);
READ32(open_conf->oc_seqid);
-
+
DECODE_TAIL;
}
@@ -991,6 +995,14 @@ nfsd4_decode_putfh(struct nfsd4_compoundargs *argp, struct nfsd4_putfh *putfh)
}
static __be32
+nfsd4_decode_putpubfh(struct nfsd4_compoundargs *argp, void *p)
+{
+ if (argp->minorversion == 0)
+ return nfs_ok;
+ return nfserr_notsupp;
+}
+
+static __be32
nfsd4_decode_read(struct nfsd4_compoundargs *argp, struct nfsd4_read *read)
{
DECODE_HEAD;
@@ -1061,6 +1073,9 @@ nfsd4_decode_renew(struct nfsd4_compoundargs *argp, clientid_t *clientid)
{
DECODE_HEAD;
+ if (argp->minorversion >= 1)
+ return nfserr_notsupp;
+
READ_BUF(sizeof(clientid_t));
COPYMEM(clientid, sizeof(clientid_t));
@@ -1111,6 +1126,9 @@ nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, struct nfsd4_setclient
{
DECODE_HEAD;
+ if (argp->minorversion >= 1)
+ return nfserr_notsupp;
+
READ_BUF(NFS4_VERIFIER_SIZE);
COPYMEM(setclientid->se_verf.data, NFS4_VERIFIER_SIZE);
@@ -1137,6 +1155,9 @@ nfsd4_decode_setclientid_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_s
{
DECODE_HEAD;
+ if (argp->minorversion >= 1)
+ return nfserr_notsupp;
+
READ_BUF(8 + NFS4_VERIFIER_SIZE);
COPYMEM(&scd_c->sc_clientid, 8);
COPYMEM(&scd_c->sc_confirm, NFS4_VERIFIER_SIZE);
@@ -1220,6 +1241,9 @@ nfsd4_decode_release_lockowner(struct nfsd4_compoundargs *argp, struct nfsd4_rel
{
DECODE_HEAD;
+ if (argp->minorversion >= 1)
+ return nfserr_notsupp;
+
READ_BUF(12);
COPYMEM(&rlockowner->rl_clientid, sizeof(clientid_t));
READ32(rlockowner->rl_owner.len);
@@ -1519,7 +1543,7 @@ static nfsd4_dec nfsd4_dec_ops[] = {
[OP_OPEN_CONFIRM] = (nfsd4_dec)nfsd4_decode_open_confirm,
[OP_OPEN_DOWNGRADE] = (nfsd4_dec)nfsd4_decode_open_downgrade,
[OP_PUTFH] = (nfsd4_dec)nfsd4_decode_putfh,
- [OP_PUTPUBFH] = (nfsd4_dec)nfsd4_decode_noop,
+ [OP_PUTPUBFH] = (nfsd4_dec)nfsd4_decode_putpubfh,
[OP_PUTROOTFH] = (nfsd4_dec)nfsd4_decode_noop,
[OP_READ] = (nfsd4_dec)nfsd4_decode_read,
[OP_READDIR] = (nfsd4_dec)nfsd4_decode_readdir,
@@ -1536,46 +1560,6 @@ static nfsd4_dec nfsd4_dec_ops[] = {
[OP_VERIFY] = (nfsd4_dec)nfsd4_decode_verify,
[OP_WRITE] = (nfsd4_dec)nfsd4_decode_write,
[OP_RELEASE_LOCKOWNER] = (nfsd4_dec)nfsd4_decode_release_lockowner,
-};
-
-static nfsd4_dec nfsd41_dec_ops[] = {
- [OP_ACCESS] = (nfsd4_dec)nfsd4_decode_access,
- [OP_CLOSE] = (nfsd4_dec)nfsd4_decode_close,
- [OP_COMMIT] = (nfsd4_dec)nfsd4_decode_commit,
- [OP_CREATE] = (nfsd4_dec)nfsd4_decode_create,
- [OP_DELEGPURGE] = (nfsd4_dec)nfsd4_decode_notsupp,
- [OP_DELEGRETURN] = (nfsd4_dec)nfsd4_decode_delegreturn,
- [OP_GETATTR] = (nfsd4_dec)nfsd4_decode_getattr,
- [OP_GETFH] = (nfsd4_dec)nfsd4_decode_noop,
- [OP_LINK] = (nfsd4_dec)nfsd4_decode_link,
- [OP_LOCK] = (nfsd4_dec)nfsd4_decode_lock,
- [OP_LOCKT] = (nfsd4_dec)nfsd4_decode_lockt,
- [OP_LOCKU] = (nfsd4_dec)nfsd4_decode_locku,
- [OP_LOOKUP] = (nfsd4_dec)nfsd4_decode_lookup,
- [OP_LOOKUPP] = (nfsd4_dec)nfsd4_decode_noop,
- [OP_NVERIFY] = (nfsd4_dec)nfsd4_decode_verify,
- [OP_OPEN] = (nfsd4_dec)nfsd4_decode_open,
- [OP_OPENATTR] = (nfsd4_dec)nfsd4_decode_notsupp,
- [OP_OPEN_CONFIRM] = (nfsd4_dec)nfsd4_decode_notsupp,
- [OP_OPEN_DOWNGRADE] = (nfsd4_dec)nfsd4_decode_open_downgrade,
- [OP_PUTFH] = (nfsd4_dec)nfsd4_decode_putfh,
- [OP_PUTPUBFH] = (nfsd4_dec)nfsd4_decode_notsupp,
- [OP_PUTROOTFH] = (nfsd4_dec)nfsd4_decode_noop,
- [OP_READ] = (nfsd4_dec)nfsd4_decode_read,
- [OP_READDIR] = (nfsd4_dec)nfsd4_decode_readdir,
- [OP_READLINK] = (nfsd4_dec)nfsd4_decode_noop,
- [OP_REMOVE] = (nfsd4_dec)nfsd4_decode_remove,
- [OP_RENAME] = (nfsd4_dec)nfsd4_decode_rename,
- [OP_RENEW] = (nfsd4_dec)nfsd4_decode_notsupp,
- [OP_RESTOREFH] = (nfsd4_dec)nfsd4_decode_noop,
- [OP_SAVEFH] = (nfsd4_dec)nfsd4_decode_noop,
- [OP_SECINFO] = (nfsd4_dec)nfsd4_decode_secinfo,
- [OP_SETATTR] = (nfsd4_dec)nfsd4_decode_setattr,
- [OP_SETCLIENTID] = (nfsd4_dec)nfsd4_decode_notsupp,
- [OP_SETCLIENTID_CONFIRM]= (nfsd4_dec)nfsd4_decode_notsupp,
- [OP_VERIFY] = (nfsd4_dec)nfsd4_decode_verify,
- [OP_WRITE] = (nfsd4_dec)nfsd4_decode_write,
- [OP_RELEASE_LOCKOWNER] = (nfsd4_dec)nfsd4_decode_notsupp,
/* new operations for NFSv4.1 */
[OP_BACKCHANNEL_CTL] = (nfsd4_dec)nfsd4_decode_backchannel_ctl,
@@ -1599,24 +1583,53 @@ static nfsd4_dec nfsd41_dec_ops[] = {
[OP_RECLAIM_COMPLETE] = (nfsd4_dec)nfsd4_decode_reclaim_complete,
};
-struct nfsd4_minorversion_ops {
- nfsd4_dec *decoders;
- int nops;
-};
+static inline bool
+nfsd4_opnum_in_range(struct nfsd4_compoundargs *argp, struct nfsd4_op *op)
+{
+ if (op->opnum < FIRST_NFS4_OP)
+ return false;
+ else if (argp->minorversion == 0 && op->opnum > LAST_NFS40_OP)
+ return false;
+ else if (argp->minorversion == 1 && op->opnum > LAST_NFS41_OP)
+ return false;
+ else if (argp->minorversion == 2 && op->opnum > LAST_NFS42_OP)
+ return false;
+ return true;
+}
-static struct nfsd4_minorversion_ops nfsd4_minorversion[] = {
- [0] = { nfsd4_dec_ops, ARRAY_SIZE(nfsd4_dec_ops) },
- [1] = { nfsd41_dec_ops, ARRAY_SIZE(nfsd41_dec_ops) },
- [2] = { nfsd41_dec_ops, ARRAY_SIZE(nfsd41_dec_ops) },
-};
+/*
+ * Return a rough estimate of the maximum possible reply size. Note the
+ * estimate includes rpc headers so is meant to be passed to
+ * svc_reserve, not svc_reserve_auth.
+ *
+ * Also note the current compound encoding permits only one operation to
+ * use pages beyond the first one, so the maximum possible length is the
+ * maximum over these values, not the sum.
+ */
+static int nfsd4_max_reply(u32 opnum)
+{
+ switch (opnum) {
+ case OP_READLINK:
+ case OP_READDIR:
+ /*
+ * Both of these ops take a single page for data and put
+ * the head and tail in another page:
+ */
+ return 2 * PAGE_SIZE;
+ case OP_READ:
+ return INT_MAX;
+ default:
+ return PAGE_SIZE;
+ }
+}
static __be32
nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
{
DECODE_HEAD;
struct nfsd4_op *op;
- struct nfsd4_minorversion_ops *ops;
bool cachethis = false;
+ int max_reply = PAGE_SIZE;
int i;
READ_BUF(4);
@@ -1640,10 +1653,9 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
}
}
- if (argp->minorversion >= ARRAY_SIZE(nfsd4_minorversion))
+ if (argp->minorversion > NFSD_SUPPORTED_MINOR_VERSION)
argp->opcnt = 0;
- ops = &nfsd4_minorversion[argp->minorversion];
for (i = 0; i < argp->opcnt; i++) {
op = &argp->ops[i];
op->replay = NULL;
@@ -1651,8 +1663,8 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
READ_BUF(4);
READ32(op->opnum);
- if (op->opnum >= FIRST_NFS4_OP && op->opnum <= LAST_NFS4_OP)
- op->status = ops->decoders[op->opnum](argp, &op->u);
+ if (nfsd4_opnum_in_range(argp, op))
+ op->status = nfsd4_dec_ops[op->opnum](argp, &op->u);
else {
op->opnum = OP_ILLEGAL;
op->status = nfserr_op_illegal;
@@ -1667,10 +1679,14 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
* op in the compound wants to be cached:
*/
cachethis |= nfsd4_cache_this_op(op);
+
+ max_reply = max(max_reply, nfsd4_max_reply(op->opnum));
}
/* Sessions make the DRC unnecessary: */
if (argp->minorversion)
cachethis = false;
+ if (max_reply != INT_MAX)
+ svc_reserve(argp->rqstp, max_reply);
argp->rqstp->rq_cachetype = cachethis ? RC_REPLBUFF : RC_NOCACHE;
DECODE_TAIL;
@@ -2375,7 +2391,7 @@ out_acl:
if (bmval0 & FATTR4_WORD0_MAXFILESIZE) {
if ((buflen -= 8) < 0)
goto out_resource;
- WRITE64(~(u64)0);
+ WRITE64(exp->ex_path.mnt->mnt_sb->s_maxbytes);
}
if (bmval0 & FATTR4_WORD0_MAXLINK) {
if ((buflen -= 4) < 0)
diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c
index 3d0e15ae6f72..3c37b160dcad 100644
--- a/fs/nfsd/nfsfh.c
+++ b/fs/nfsd/nfsfh.c
@@ -598,22 +598,20 @@ fh_update(struct svc_fh *fhp)
_fh_update_old(dentry, fhp->fh_export, &fhp->fh_handle);
} else {
if (fhp->fh_handle.fh_fileid_type != FILEID_ROOT)
- goto out;
+ return 0;
_fh_update(fhp, fhp->fh_export, dentry);
if (fhp->fh_handle.fh_fileid_type == FILEID_INVALID)
return nfserr_opnotsupp;
}
-out:
return 0;
-
out_bad:
printk(KERN_ERR "fh_update: fh not verified!\n");
- goto out;
+ return nfserr_serverfault;
out_negative:
printk(KERN_ERR "fh_update: %pd2 still negative!\n",
dentry);
- goto out;
+ return nfserr_serverfault;
}
/*