diff options
Diffstat (limited to 'net/sunrpc')
28 files changed, 720 insertions, 342 deletions
diff --git a/net/sunrpc/Kconfig b/net/sunrpc/Kconfig index 03d03e37a7d..241b54f3020 100644 --- a/net/sunrpc/Kconfig +++ b/net/sunrpc/Kconfig @@ -3,6 +3,7 @@ config SUNRPC config SUNRPC_GSS tristate + select OID_REGISTRY config SUNRPC_BACKCHANNEL bool @@ -10,7 +11,7 @@ config SUNRPC_BACKCHANNEL config SUNRPC_XPRT_RDMA tristate - depends on SUNRPC && INFINIBAND && INFINIBAND_ADDR_TRANS && EXPERIMENTAL + depends on SUNRPC && INFINIBAND && INFINIBAND_ADDR_TRANS default SUNRPC && INFINIBAND help This option allows the NFS client and server to support @@ -24,7 +25,6 @@ config SUNRPC_XPRT_RDMA config SUNRPC_SWAP bool depends on SUNRPC - select NETVM config RPCSEC_GSS_KRB5 tristate "Secure RPC: Kerberos V mechanism" diff --git a/net/sunrpc/addr.c b/net/sunrpc/addr.c index d11418f97f1..a622ad64acd 100644 --- a/net/sunrpc/addr.c +++ b/net/sunrpc/addr.c @@ -17,7 +17,8 @@ */ #include <net/ipv6.h> -#include <linux/sunrpc/clnt.h> +#include <linux/sunrpc/addr.h> +#include <linux/sunrpc/msg_prot.h> #include <linux/slab.h> #include <linux/export.h> diff --git a/net/sunrpc/auth.c b/net/sunrpc/auth.c index b5c067bccc4..ed2fdd210c0 100644 --- a/net/sunrpc/auth.c +++ b/net/sunrpc/auth.c @@ -82,7 +82,7 @@ MODULE_PARM_DESC(auth_hashtable_size, "RPC credential cache hashtable size"); static u32 pseudoflavor_to_flavor(u32 flavor) { - if (flavor >= RPC_AUTH_MAXFLAVOR) + if (flavor > RPC_AUTH_MAXFLAVOR) return RPC_AUTH_GSS; return flavor; } @@ -124,6 +124,79 @@ rpcauth_unregister(const struct rpc_authops *ops) EXPORT_SYMBOL_GPL(rpcauth_unregister); /** + * rpcauth_get_pseudoflavor - check if security flavor is supported + * @flavor: a security flavor + * @info: a GSS mech OID, quality of protection, and service value + * + * Verifies that an appropriate kernel module is available or already loaded. + * Returns an equivalent pseudoflavor, or RPC_AUTH_MAXFLAVOR if "flavor" is + * not supported locally. + */ +rpc_authflavor_t +rpcauth_get_pseudoflavor(rpc_authflavor_t flavor, struct rpcsec_gss_info *info) +{ + const struct rpc_authops *ops; + rpc_authflavor_t pseudoflavor; + + ops = auth_flavors[flavor]; + if (ops == NULL) + request_module("rpc-auth-%u", flavor); + spin_lock(&rpc_authflavor_lock); + ops = auth_flavors[flavor]; + if (ops == NULL || !try_module_get(ops->owner)) { + spin_unlock(&rpc_authflavor_lock); + return RPC_AUTH_MAXFLAVOR; + } + spin_unlock(&rpc_authflavor_lock); + + pseudoflavor = flavor; + if (ops->info2flavor != NULL) + pseudoflavor = ops->info2flavor(info); + + module_put(ops->owner); + return pseudoflavor; +} +EXPORT_SYMBOL_GPL(rpcauth_get_pseudoflavor); + +/** + * rpcauth_get_gssinfo - find GSS tuple matching a GSS pseudoflavor + * @pseudoflavor: GSS pseudoflavor to match + * @info: rpcsec_gss_info structure to fill in + * + * Returns zero and fills in "info" if pseudoflavor matches a + * supported mechanism. + */ +int +rpcauth_get_gssinfo(rpc_authflavor_t pseudoflavor, struct rpcsec_gss_info *info) +{ + rpc_authflavor_t flavor = pseudoflavor_to_flavor(pseudoflavor); + const struct rpc_authops *ops; + int result; + + if (flavor >= RPC_AUTH_MAXFLAVOR) + return -EINVAL; + + ops = auth_flavors[flavor]; + if (ops == NULL) + request_module("rpc-auth-%u", flavor); + spin_lock(&rpc_authflavor_lock); + ops = auth_flavors[flavor]; + if (ops == NULL || !try_module_get(ops->owner)) { + spin_unlock(&rpc_authflavor_lock); + return -ENOENT; + } + spin_unlock(&rpc_authflavor_lock); + + result = -ENOENT; + if (ops->flavor2info != NULL) + result = ops->flavor2info(pseudoflavor, info); + + module_put(ops->owner); + return result; +} +EXPORT_SYMBOL_GPL(rpcauth_get_gssinfo); + +/** * rpcauth_list_flavors - discover registered flavors and pseudoflavors * @array: array to fill in * @size: size of "array" @@ -407,15 +480,14 @@ rpcauth_lookup_credcache(struct rpc_auth *auth, struct auth_cred * acred, { LIST_HEAD(free); struct rpc_cred_cache *cache = auth->au_credcache; - struct hlist_node *pos; struct rpc_cred *cred = NULL, *entry, *new; unsigned int nr; - nr = hash_long(acred->uid, cache->hashbits); + nr = hash_long(from_kuid(&init_user_ns, acred->uid), cache->hashbits); rcu_read_lock(); - hlist_for_each_entry_rcu(entry, pos, &cache->hashtable[nr], cr_hash) { + hlist_for_each_entry_rcu(entry, &cache->hashtable[nr], cr_hash) { if (!entry->cr_ops->crmatch(acred, entry, flags)) continue; spin_lock(&cache->lock); @@ -439,7 +511,7 @@ rpcauth_lookup_credcache(struct rpc_auth *auth, struct auth_cred * acred, } spin_lock(&cache->lock); - hlist_for_each_entry(entry, pos, &cache->hashtable[nr], cr_hash) { + hlist_for_each_entry(entry, &cache->hashtable[nr], cr_hash) { if (!entry->cr_ops->crmatch(acred, entry, flags)) continue; cred = get_rpccred(entry); @@ -519,8 +591,8 @@ rpcauth_bind_root_cred(struct rpc_task *task, int lookupflags) { struct rpc_auth *auth = task->tk_client->cl_auth; struct auth_cred acred = { - .uid = 0, - .gid = 0, + .uid = GLOBAL_ROOT_UID, + .gid = GLOBAL_ROOT_GID, }; dprintk("RPC: %5u looking up %s cred\n", diff --git a/net/sunrpc/auth_generic.c b/net/sunrpc/auth_generic.c index 6ed6f201b02..b6badafc649 100644 --- a/net/sunrpc/auth_generic.c +++ b/net/sunrpc/auth_generic.c @@ -18,8 +18,8 @@ # define RPCDBG_FACILITY RPCDBG_AUTH #endif -#define RPC_MACHINE_CRED_USERID ((uid_t)0) -#define RPC_MACHINE_CRED_GROUPID ((gid_t)0) +#define RPC_MACHINE_CRED_USERID GLOBAL_ROOT_UID +#define RPC_MACHINE_CRED_GROUPID GLOBAL_ROOT_GID struct generic_cred { struct rpc_cred gc_base; @@ -96,7 +96,9 @@ generic_create_cred(struct rpc_auth *auth, struct auth_cred *acred, int flags) dprintk("RPC: allocated %s cred %p for uid %d gid %d\n", gcred->acred.machine_cred ? "machine" : "generic", - gcred, acred->uid, acred->gid); + gcred, + from_kuid(&init_user_ns, acred->uid), + from_kgid(&init_user_ns, acred->gid)); return &gcred->gc_base; } @@ -129,8 +131,8 @@ machine_cred_match(struct auth_cred *acred, struct generic_cred *gcred, int flag { if (!gcred->acred.machine_cred || gcred->acred.principal != acred->principal || - gcred->acred.uid != acred->uid || - gcred->acred.gid != acred->gid) + !uid_eq(gcred->acred.uid, acred->uid) || + !gid_eq(gcred->acred.gid, acred->gid)) return 0; return 1; } @@ -147,8 +149,8 @@ generic_match(struct auth_cred *acred, struct rpc_cred *cred, int flags) if (acred->machine_cred) return machine_cred_match(acred, gcred, flags); - if (gcred->acred.uid != acred->uid || - gcred->acred.gid != acred->gid || + if (!uid_eq(gcred->acred.uid, acred->uid) || + !gid_eq(gcred->acred.gid, acred->gid) || gcred->acred.machine_cred != 0) goto out_nomatch; diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c index 6e5c824b040..51415b07174 100644 --- a/net/sunrpc/auth_gss/auth_gss.c +++ b/net/sunrpc/auth_gss/auth_gss.c @@ -247,8 +247,7 @@ gss_fill_context(const void *p, const void *end, struct gss_cl_ctx *ctx, struct __func__, ctx->gc_expiry, now, timeout); return q; err: - dprintk("RPC: %s returns %ld gc_expiry %lu now %lu timeout %u\n", - __func__, -PTR_ERR(p), ctx->gc_expiry, now, timeout); + dprintk("RPC: %s returns error %ld\n", __func__, -PTR_ERR(p)); return p; } @@ -256,7 +255,7 @@ err: struct gss_upcall_msg { atomic_t count; - uid_t uid; + kuid_t uid; struct rpc_pipe_msg msg; struct list_head list; struct gss_auth *auth; @@ -303,11 +302,11 @@ gss_release_msg(struct gss_upcall_msg *gss_msg) } static struct gss_upcall_msg * -__gss_find_upcall(struct rpc_pipe *pipe, uid_t uid) +__gss_find_upcall(struct rpc_pipe *pipe, kuid_t uid) { struct gss_upcall_msg *pos; list_for_each_entry(pos, &pipe->in_downcall, list) { - if (pos->uid != uid) + if (!uid_eq(pos->uid, uid)) continue; atomic_inc(&pos->count); dprintk("RPC: %s found msg %p\n", __func__, pos); @@ -395,8 +394,11 @@ gss_upcall_callback(struct rpc_task *task) static void gss_encode_v0_msg(struct gss_upcall_msg *gss_msg) { - gss_msg->msg.data = &gss_msg->uid; - gss_msg->msg.len = sizeof(gss_msg->uid); + uid_t uid = from_kuid(&init_user_ns, gss_msg->uid); + memcpy(gss_msg->databuf, &uid, sizeof(uid)); + gss_msg->msg.data = gss_msg->databuf; + gss_msg->msg.len = sizeof(uid); + BUG_ON(sizeof(uid) > UPCALL_BUF_LEN); } static void gss_encode_v1_msg(struct gss_upcall_msg *gss_msg, @@ -409,7 +411,7 @@ static void gss_encode_v1_msg(struct gss_upcall_msg *gss_msg, gss_msg->msg.len = sprintf(gss_msg->databuf, "mech=%s uid=%d ", mech->gm_name, - gss_msg->uid); + from_kuid(&init_user_ns, gss_msg->uid)); p += gss_msg->msg.len; if (clnt->cl_principal) { len = sprintf(p, "target=%s ", clnt->cl_principal); @@ -445,7 +447,7 @@ static void gss_encode_msg(struct gss_upcall_msg *gss_msg, static struct gss_upcall_msg * gss_alloc_msg(struct gss_auth *gss_auth, struct rpc_clnt *clnt, - uid_t uid, const char *service_name) + kuid_t uid, const char *service_name) { struct gss_upcall_msg *gss_msg; int vers; @@ -475,7 +477,7 @@ gss_setup_upcall(struct rpc_clnt *clnt, struct gss_auth *gss_auth, struct rpc_cr struct gss_cred *gss_cred = container_of(cred, struct gss_cred, gc_base); struct gss_upcall_msg *gss_new, *gss_msg; - uid_t uid = cred->cr_uid; + kuid_t uid = cred->cr_uid; gss_new = gss_alloc_msg(gss_auth, clnt, uid, gss_cred->gc_principal); if (IS_ERR(gss_new)) @@ -517,7 +519,7 @@ gss_refresh_upcall(struct rpc_task *task) int err = 0; dprintk("RPC: %5u %s for uid %u\n", - task->tk_pid, __func__, cred->cr_uid); + task->tk_pid, __func__, from_kuid(&init_user_ns, cred->cr_uid)); gss_msg = gss_setup_upcall(task->tk_client, gss_auth, cred); if (PTR_ERR(gss_msg) == -EAGAIN) { /* XXX: warning on the first, under the assumption we @@ -549,7 +551,8 @@ gss_refresh_upcall(struct rpc_task *task) gss_release_msg(gss_msg); out: dprintk("RPC: %5u %s for uid %u result %d\n", - task->tk_pid, __func__, cred->cr_uid, err); + task->tk_pid, __func__, + from_kuid(&init_user_ns, cred->cr_uid), err); return err; } @@ -562,7 +565,8 @@ gss_create_upcall(struct gss_auth *gss_auth, struct gss_cred *gss_cred) DEFINE_WAIT(wait); int err = 0; - dprintk("RPC: %s for uid %u\n", __func__, cred->cr_uid); + dprintk("RPC: %s for uid %u\n", + __func__, from_kuid(&init_user_ns, cred->cr_uid)); retry: gss_msg = gss_setup_upcall(gss_auth->client, gss_auth, cred); if (PTR_ERR(gss_msg) == -EAGAIN) { @@ -604,7 +608,7 @@ out_intr: gss_release_msg(gss_msg); out: dprintk("RPC: %s for uid %u result %d\n", - __func__, cred->cr_uid, err); + __func__, from_kuid(&init_user_ns, cred->cr_uid), err); return err; } @@ -616,9 +620,10 @@ gss_pipe_downcall(struct file *filp, const char __user *src, size_t mlen) const void *p, *end; void *buf; struct gss_upcall_msg *gss_msg; - struct rpc_pipe *pipe = RPC_I(filp->f_dentry->d_inode)->pipe; + struct rpc_pipe *pipe = RPC_I(file_inode(filp))->pipe; struct gss_cl_ctx *ctx; - uid_t uid; + uid_t id; + kuid_t uid; ssize_t err = -EFBIG; if (mlen > MSG_BUF_MAXSIZE) @@ -633,12 +638,18 @@ gss_pipe_downcall(struct file *filp, const char __user *src, size_t mlen) goto err; end = (const void *)((char *)buf + mlen); - p = simple_get_bytes(buf, end, &uid, sizeof(uid)); + p = simple_get_bytes(buf, end, &id, sizeof(id)); if (IS_ERR(p)) { err = PTR_ERR(p); goto err; } + uid = make_kuid(&init_user_ns, id); + if (!uid_valid(uid)) { + err = -EINVAL; + goto err; + } + err = -ENOMEM; ctx = gss_alloc_context(); if (ctx == NULL) @@ -1059,7 +1070,8 @@ gss_create_cred(struct rpc_auth *auth, struct auth_cred *acred, int flags) int err = -ENOMEM; dprintk("RPC: %s for uid %d, flavor %d\n", - __func__, acred->uid, auth->au_flavor); + __func__, from_kuid(&init_user_ns, acred->uid), + auth->au_flavor); if (!(cred = kzalloc(sizeof(*cred), GFP_NOFS))) goto out_err; @@ -1115,7 +1127,7 @@ out: } if (gss_cred->gc_principal != NULL) return 0; - return rc->cr_uid == acred->uid; + return uid_eq(rc->cr_uid, acred->uid); } /* @@ -1154,7 +1166,7 @@ gss_marshal(struct rpc_task *task, __be32 *p) /* We compute the checksum for the verifier over the xdr-encoded bytes * starting with the xid and ending at the end of the credential: */ - iov.iov_base = xprt_skip_transport_header(task->tk_xprt, + iov.iov_base = xprt_skip_transport_header(req->rq_xprt, req->rq_snd_buf.head[0].iov_base); iov.iov_len = (u8 *)p - (u8 *)iov.iov_base; xdr_buf_from_iov(&iov, &verf_buf); @@ -1629,6 +1641,8 @@ static const struct rpc_authops authgss_ops = { .pipes_create = gss_pipes_dentries_create, .pipes_destroy = gss_pipes_dentries_destroy, .list_pseudoflavors = gss_mech_list_pseudoflavors, + .info2flavor = gss_mech_info2flavor, + .flavor2info = gss_mech_flavor2info, }; static const struct rpc_credops gss_credops = { @@ -1721,6 +1735,7 @@ static void __exit exit_rpcsec_gss(void) rcu_barrier(); /* Wait for completion of call_rcu()'s */ } +MODULE_ALIAS("rpc-auth-6"); MODULE_LICENSE("GPL"); module_param_named(expired_cred_retry_delay, gss_expired_cred_retry_delay, diff --git a/net/sunrpc/auth_gss/gss_krb5_mech.c b/net/sunrpc/auth_gss/gss_krb5_mech.c index d3611f11a8d..33255ff889c 100644 --- a/net/sunrpc/auth_gss/gss_krb5_mech.c +++ b/net/sunrpc/auth_gss/gss_krb5_mech.c @@ -729,16 +729,19 @@ static const struct gss_api_ops gss_kerberos_ops = { static struct pf_desc gss_kerberos_pfs[] = { [0] = { .pseudoflavor = RPC_AUTH_GSS_KRB5, + .qop = GSS_C_QOP_DEFAULT, .service = RPC_GSS_SVC_NONE, .name = "krb5", }, [1] = { .pseudoflavor = RPC_AUTH_GSS_KRB5I, + .qop = GSS_C_QOP_DEFAULT, .service = RPC_GSS_SVC_INTEGRITY, .name = "krb5i", }, [2] = { .pseudoflavor = RPC_AUTH_GSS_KRB5P, + .qop = GSS_C_QOP_DEFAULT, .service = RPC_GSS_SVC_PRIVACY, .name = "krb5p", }, @@ -750,11 +753,12 @@ MODULE_ALIAS("rpc-auth-gss-krb5p"); MODULE_ALIAS("rpc-auth-gss-390003"); MODULE_ALIAS("rpc-auth-gss-390004"); MODULE_ALIAS("rpc-auth-gss-390005"); +MODULE_ALIAS("rpc-auth-gss-1.2.840.113554.1.2.2"); static struct gss_api_mech gss_kerberos_mech = { .gm_name = "krb5", .gm_owner = THIS_MODULE, - .gm_oid = {9, (void *)"\x2a\x86\x48\x86\xf7\x12\x01\x02\x02"}, + .gm_oid = { 9, "\x2a\x86\x48\x86\xf7\x12\x01\x02\x02" }, .gm_ops = &gss_kerberos_ops, .gm_pf_num = ARRAY_SIZE(gss_kerberos_pfs), .gm_pfs = gss_kerberos_pfs, diff --git a/net/sunrpc/auth_gss/gss_krb5_wrap.c b/net/sunrpc/auth_gss/gss_krb5_wrap.c index 107c4528654..1da52d1406f 100644 --- a/net/sunrpc/auth_gss/gss_krb5_wrap.c +++ b/net/sunrpc/auth_gss/gss_krb5_wrap.c @@ -130,8 +130,8 @@ gss_krb5_make_confounder(char *p, u32 conflen) /* initialize to random value */ if (i == 0) { - i = random32(); - i = (i << 32) | random32(); + i = prandom_u32(); + i = (i << 32) | prandom_u32(); } switch (conflen) { @@ -574,6 +574,8 @@ gss_unwrap_kerberos_v2(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf) buf->head[0].iov_len -= GSS_KRB5_TOK_HDR_LEN + headskip; buf->len -= GSS_KRB5_TOK_HDR_LEN + headskip; + /* Trim off the checksum blob */ + xdr_buf_trim(buf, GSS_KRB5_TOK_HDR_LEN + tailskip); return GSS_S_COMPLETE; } diff --git a/net/sunrpc/auth_gss/gss_mech_switch.c b/net/sunrpc/auth_gss/gss_mech_switch.c index b174fcd9ff4..79881d6e68a 100644 --- a/net/sunrpc/auth_gss/gss_mech_switch.c +++ b/net/sunrpc/auth_gss/gss_mech_switch.c @@ -36,6 +36,7 @@ #include <linux/types.h> #include <linux/slab.h> #include <linux/module.h> +#include <linux/oid_registry.h> #include <linux/sunrpc/msg_prot.h> #include <linux/sunrpc/gss_asn1.h> #include <linux/sunrpc/auth_gss.h> @@ -102,8 +103,13 @@ out: return status; } -int -gss_mech_register(struct gss_api_mech *gm) +/** + * gss_mech_register - register a GSS mechanism + * @gm: GSS mechanism handle + * + * Returns zero if successful, or a negative errno. + */ +int gss_mech_register(struct gss_api_mech *gm) { int status; @@ -116,11 +122,14 @@ gss_mech_register(struct gss_api_mech *gm) dprintk("RPC: registered gss mechanism %s\n", gm->gm_name); return 0; } - EXPORT_SYMBOL_GPL(gss_mech_register); -void -gss_mech_unregister(struct gss_api_mech *gm) +/** + * gss_mech_unregister - release a GSS mechanism + * @gm: GSS mechanism handle + * + */ +void gss_mech_unregister(struct gss_api_mech *gm) { spin_lock(®istered_mechs_lock); list_del(&gm->gm_list); @@ -128,19 +137,15 @@ gss_mech_unregister(struct gss_api_mech *gm) dprintk("RPC: unregistered gss mechanism %s\n", gm->gm_name); gss_mech_free(gm); } - EXPORT_SYMBOL_GPL(gss_mech_unregister); -struct gss_api_mech * -gss_mech_get(struct gss_api_mech *gm) +static struct gss_api_mech *gss_mech_get(struct gss_api_mech *gm) { __module_get(gm->gm_owner); return gm; } -EXPORT_SYMBOL_GPL(gss_mech_get); - -struct gss_api_mech * +static struct gss_api_mech * _gss_mech_get_by_name(const char *name) { struct gss_api_mech *pos, *gm = NULL; @@ -169,12 +174,16 @@ struct gss_api_mech * gss_mech_get_by_name(const char *name) } return gm; } -EXPORT_SYMBOL_GPL(gss_mech_get_by_name); -struct gss_api_mech * -gss_mech_get_by_OID(struct xdr_netobj *obj) +static struct gss_api_mech *gss_mech_get_by_OID(struct rpcsec_gss_oid *obj) { struct gss_api_mech *pos, *gm = NULL; + char buf[32]; + + if (sprint_oid(obj->data, obj->len, buf, sizeof(buf)) < 0) + return NULL; + dprintk("RPC: %s(%s)\n", __func__, buf); + request_module("rpc-auth-gss-%s", buf); spin_lock(®istered_mechs_lock); list_for_each_entry(pos, ®istered_mechs, gm_list) { @@ -188,11 +197,8 @@ gss_mech_get_by_OID(struct xdr_netobj *obj) } spin_unlock(®istered_mechs_lock); return gm; - } -EXPORT_SYMBOL_GPL(gss_mech_get_by_OID); - static inline int mech_supports_pseudoflavor(struct gss_api_mech *gm, u32 pseudoflavor) { @@ -205,7 +211,7 @@ mech_supports_pseudoflavor(struct gss_api_mech *gm, u32 pseudoflavor) return 0; } -struct gss_api_mech *_gss_mech_get_by_pseudoflavor(u32 pseudoflavor) +static struct gss_api_mech *_gss_mech_get_by_pseudoflavor(u32 pseudoflavor) { struct gss_api_mech *gm = NULL, *pos; @@ -237,8 +243,6 @@ gss_mech_get_by_pseudoflavor(u32 pseudoflavor) return gm; } -EXPORT_SYMBOL_GPL(gss_mech_get_by_pseudoflavor); - /** * gss_mech_list_pseudoflavors - Discover registered GSS pseudoflavors * @array: array to fill in @@ -268,19 +272,82 @@ int gss_mech_list_pseudoflavors(rpc_authflavor_t *array_ptr, int size) return i; } -u32 -gss_svc_to_pseudoflavor(struct gss_api_mech *gm, u32 service) +/** + * gss_svc_to_pseudoflavor - map a GSS service number to a pseudoflavor + * @gm: GSS mechanism handle + * @qop: GSS quality-of-protection value + * @service: GSS service value + * + * Returns a matching security flavor, or RPC_AUTH_MAXFLAVOR if none is found. + */ +rpc_authflavor_t gss_svc_to_pseudoflavor(struct gss_api_mech *gm, u32 qop, + u32 service) { int i; for (i = 0; i < gm->gm_pf_num; i++) { - if (gm->gm_pfs[i].service == service) { + if (gm->gm_pfs[i].qop == qop && + gm->gm_pfs[i].service == service) { return gm->gm_pfs[i].pseudoflavor; } } - return RPC_AUTH_MAXFLAVOR; /* illegal value */ + return RPC_AUTH_MAXFLAVOR; +} + +/** + * gss_mech_info2flavor - look up a pseudoflavor given a GSS tuple + * @info: a GSS mech OID, quality of protection, and service value + * + * Returns a matching pseudoflavor, or RPC_AUTH_MAXFLAVOR if the tuple is + * not supported. + */ +rpc_authflavor_t gss_mech_info2flavor(struct rpcsec_gss_info *info) +{ + rpc_authflavor_t pseudoflavor; + struct gss_api_mech *gm; + + gm = gss_mech_get_by_OID(&info->oid); + if (gm == NULL) + return RPC_AUTH_MAXFLAVOR; + + pseudoflavor = gss_svc_to_pseudoflavor(gm, info->qop, info->service); + + gss_mech_put(gm); + return pseudoflavor; +} + +/** + * gss_mech_flavor2info - look up a GSS tuple for a given pseudoflavor + * @pseudoflavor: GSS pseudoflavor to match + * @info: rpcsec_gss_info structure to fill in + * + * Returns zero and fills in "info" if pseudoflavor matches a + * supported mechanism. Otherwise a negative errno is returned. + */ +int gss_mech_flavor2info(rpc_authflavor_t pseudoflavor, + struct rpcsec_gss_info *info) +{ + struct gss_api_mech *gm; + int i; + + gm = gss_mech_get_by_pseudoflavor(pseudoflavor); + if (gm == NULL) + return -ENOENT; + + for (i = 0; i < gm->gm_pf_num; i++) { + if (gm->gm_pfs[i].pseudoflavor == pseudoflavor) { + memcpy(info->oid.data, gm->gm_oid.data, gm->gm_oid.len); + info->oid.len = gm->gm_oid.len; + info->qop = gm->gm_pfs[i].qop; + info->service = gm->gm_pfs[i].service; + gss_mech_put(gm); + return 0; + } + } + + gss_mech_put(gm); + return -ENOENT; } -EXPORT_SYMBOL_GPL(gss_svc_to_pseudoflavor); u32 gss_pseudoflavor_to_service(struct gss_api_mech *gm, u32 pseudoflavor) @@ -294,8 +361,6 @@ gss_pseudoflavor_to_service(struct gss_api_mech *gm, u32 pseudoflavor) return 0; } -EXPORT_SYMBOL_GPL(gss_pseudoflavor_to_service); - char * gss_service_to_auth_domain_name(struct gss_api_mech *gm, u32 service) { @@ -308,8 +373,6 @@ gss_service_to_auth_domain_name(struct gss_api_mech *gm, u32 service) return NULL; } -EXPORT_SYMBOL_GPL(gss_service_to_auth_domain_name); - void gss_mech_put(struct gss_api_mech * gm) { @@ -317,8 +380,6 @@ gss_mech_put(struct gss_api_mech * gm) module_put(gm->gm_owner); } -EXPORT_SYMBOL_GPL(gss_mech_put); - /* The mech could probably be determined from the token instead, but it's just * as easy for now to pass it in. */ int diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c index 73e95738660..c3ba570222d 100644 --- a/net/sunrpc/auth_gss/svcauth_gss.c +++ b/net/sunrpc/auth_gss/svcauth_gss.c @@ -182,12 +182,6 @@ static void rsi_request(struct cache_detail *cd, (*bpp)[-1] = '\n'; } -static int rsi_upcall(struct cache_detail *cd, struct cache_head *h) -{ - return sunrpc_cache_pipe_upcall(cd, h, rsi_request); -} - - static int rsi_parse(struct cache_detail *cd, char *mesg, int mlen) { @@ -275,7 +269,7 @@ static struct cache_detail rsi_cache_template = { .hash_size = RSI_HASHMAX, .name = "auth.rpcsec.init", .cache_put = rsi_put, - .cache_upcall = rsi_upcall, + .cache_request = rsi_request, .cache_parse = rsi_parse, .match = rsi_match, .init = rsi_init, @@ -418,6 +412,7 @@ static int rsc_parse(struct cache_detail *cd, { /* contexthandle expiry [ uid gid N <n gids> mechname ...mechdata... ] */ char *buf = mesg; + int id; int len, rv; struct rsc rsci, *rscp = NULL; time_t expiry; @@ -444,7 +439,7 @@ static int rsc_parse(struct cache_detail *cd, goto out; /* uid, or NEGATIVE */ - rv = get_int(&mesg, &rsci.cred.cr_uid); + rv = get_int(&mesg, &id); if (rv == -EINVAL) goto out; if (rv == -ENOENT) @@ -452,9 +447,21 @@ static int rsc_parse(struct cache_detail *cd, else { int N, i; + /* + * NOTE: we skip uid_valid()/gid_valid() checks here: + * instead, * -1 id's are later mapped to the + * (export-specific) anonymous id by nfsd_setuser. + * + * (But supplementary gid's get no such special + * treatment so are checked for validity here.) + */ + /* uid */ + rsci.cred.cr_uid = make_kuid(&init_user_ns, id); + /* gid */ - if (get_int(&mesg, &rsci.cred.cr_gid)) + if (get_int(&mesg, &id)) goto out; + rsci.cred.cr_gid = make_kgid(&init_user_ns, id); /* number of additional gid's */ if (get_int(&mesg, &N)) @@ -467,11 +474,10 @@ static int rsc_parse(struct cache_detail *cd, /* gid's */ status = -EINVAL; for (i=0; i<N; i++) { - gid_t gid; kgid_t kgid; - if (get_int(&mesg, &gid)) + if (get_int(&mesg, &id)) goto out; - kgid = make_kgid(&init_user_ns, gid); + kgid = make_kgid(&init_user_ns, id); if (!gid_valid(kgid)) goto out; GROUP_AT(rsci.cred.cr_group_info, i) = kgid; @@ -817,13 +823,17 @@ read_u32_from_xdr_buf(struct xdr_buf *buf, int base, u32 *obj) * The server uses base of head iovec as read pointer, while the * client uses separate pointer. */ static int -unwrap_integ_data(struct xdr_buf *buf, u32 seq, struct gss_ctx *ctx) +unwrap_integ_data(struct svc_rqst *rqstp, struct xdr_buf *buf, u32 seq, struct gss_ctx *ctx) { int stat = -EINVAL; u32 integ_len, maj_stat; struct xdr_netobj mic; struct xdr_buf integ_buf; + /* Did we already verify the signature on the original pass through? */ + if (rqstp->rq_deferred) + return 0; + integ_len = svc_getnl(&buf->head[0]); if (integ_len & 3) return stat; @@ -846,6 +856,8 @@ unwrap_integ_data(struct xdr_buf *buf, u32 seq, struct gss_ctx *ctx) goto out; if (svc_getnl(&buf->head[0]) != seq) goto out; + /* trim off the mic at the end before returning */ + xdr_buf_trim(buf, mic.len + 4); stat = 0; out: kfree(mic.data); @@ -1190,7 +1202,7 @@ svcauth_gss_accept(struct svc_rqst *rqstp, __be32 *authp) /* placeholders for length and seq. number: */ svc_putnl(resv, 0); svc_putnl(resv, 0); - if (unwrap_integ_data(&rqstp->rq_arg, + if (unwrap_integ_data(rqstp, &rqstp->rq_arg, gc->gc_seq, rsci->mechctx)) goto garbage_args; break; @@ -1208,7 +1220,9 @@ svcauth_gss_accept(struct svc_rqst *rqstp, __be32 *authp) svcdata->rsci = rsci; cache_get(&rsci->h); rqstp->rq_cred.cr_flavor = gss_svc_to_pseudoflavor( - rsci->mechctx->mech_type, gc->gc_svc); + rsci->mechctx->mech_type, + GSS_C_QOP_DEFAULT, + gc->gc_svc); ret = SVC_OK; goto out; } diff --git a/net/sunrpc/auth_unix.c b/net/sunrpc/auth_unix.c index 52c5abdee21..dc37021fc3e 100644 --- a/net/sunrpc/auth_unix.c +++ b/net/sunrpc/auth_unix.c @@ -18,8 +18,8 @@ struct unx_cred { struct rpc_cred uc_base; - gid_t uc_gid; - gid_t uc_gids[NFS_NGROUPS]; + kgid_t uc_gid; + kgid_t uc_gids[NFS_NGROUPS]; }; #define uc_uid uc_base.cr_uid @@ -65,7 +65,8 @@ unx_create_cred(struct rpc_auth *auth, struct auth_cred *acred, int flags) unsigned int i; dprintk("RPC: allocating UNIX cred for uid %d gid %d\n", - acred->uid, acred->gid); + from_kuid(&init_user_ns, acred->uid), + from_kgid(&init_user_ns, acred->gid)); if (!(cred = kmalloc(sizeof(*cred), GFP_NOFS))) return ERR_PTR(-ENOMEM); @@ -79,13 +80,10 @@ unx_create_cred(struct rpc_auth *auth, struct auth_cred *acred, int flags) groups = NFS_NGROUPS; cred->uc_gid = acred->gid; - for (i = 0; i < groups; i++) { - gid_t gid; - gid = from_kgid(&init_user_ns, GROUP_AT(acred->group_info, i)); - cred->uc_gids[i] = gid; - } + for (i = 0; i < groups; i++) + cred->uc_gids[i] = GROUP_AT(acred->group_info, i); if (i < NFS_NGROUPS) - cred->uc_gids[i] = NOGROUP; + cred->uc_gids[i] = INVALID_GID; return &cred->uc_base; } @@ -123,21 +121,17 @@ unx_match(struct auth_cred *acred, struct rpc_cred *rcred, int flags) unsigned int i; - if (cred->uc_uid != acred->uid || cred->uc_gid != acred->gid) + if (!uid_eq(cred->uc_uid, acred->uid) || !gid_eq(cred->uc_gid, acred->gid)) return 0; if (acred->group_info != NULL) groups = acred->group_info->ngroups; if (groups > NFS_NGROUPS) groups = NFS_NGROUPS; - for (i = 0; i < groups ; i++) { - gid_t gid; - gid = from_kgid(&init_user_ns, GROUP_AT(acred->group_info, i)); - if (cred->uc_gids[i] != gid) + for (i = 0; i < groups ; i++) + if (!gid_eq(cred->uc_gids[i], GROUP_AT(acred->group_info, i))) return 0; - } - if (groups < NFS_NGROUPS && - cred->uc_gids[groups] != NOGROUP) + if (groups < NFS_NGROUPS && gid_valid(cred->uc_gids[groups])) return 0; return 1; } @@ -163,11 +157,11 @@ unx_marshal(struct rpc_task *task, __be32 *p) */ p = xdr_encode_array(p, clnt->cl_nodename, clnt->cl_nodelen); - *p++ = htonl((u32) cred->uc_uid); - *p++ = htonl((u32) cred->uc_gid); + *p++ = htonl((u32) from_kuid(&init_user_ns, cred->uc_uid)); + *p++ = htonl((u32) from_kgid(&init_user_ns, cred->uc_gid)); hold = p++; - for (i = 0; i < 16 && cred->uc_gids[i] != (gid_t) NOGROUP; i++) - *p++ = htonl((u32) cred->uc_gids[i]); + for (i = 0; i < 16 && gid_valid(cred->uc_gids[i]); i++) + *p++ = htonl((u32) from_kgid(&init_user_ns, cred->uc_gids[i])); *hold = htonl(p - hold - 1); /* gid array length */ *base = htonl((p - base - 1) << 2); /* cred length */ diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index 9afa4393c21..f1889be8091 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -196,9 +196,9 @@ EXPORT_SYMBOL_GPL(sunrpc_cache_update); static int cache_make_upcall(struct cache_detail *cd, struct cache_head *h) { - if (!cd->cache_upcall) - return -EINVAL; - return cd->cache_upcall(cd, h); + if (cd->cache_upcall) + return cd->cache_upcall(cd, h); + return sunrpc_cache_pipe_upcall(cd, h); } static inline int cache_is_valid(struct cache_detail *detail, struct cache_head *h) @@ -670,13 +670,13 @@ static void cache_revisit_request(struct cache_head *item) { struct cache_deferred_req *dreq; struct list_head pending; - struct hlist_node *lp, *tmp; + struct hlist_node *tmp; int hash = DFR_HASH(item); INIT_LIST_HEAD(&pending); spin_lock(&cache_defer_lock); - hlist_for_each_entry_safe(dreq, lp, tmp, &cache_defer_hash[hash], hash) + hlist_for_each_entry_safe(dreq, tmp, &cache_defer_hash[hash], hash) if (dreq->item == item) { __unhash_deferred_req(dreq); list_add(&dreq->recent, &pending); @@ -750,12 +750,24 @@ struct cache_reader { int offset; /* if non-0, we have a refcnt on next request */ }; +static int cache_request(struct cache_detail *detail, + struct cache_request *crq) +{ + char *bp = crq->buf; + int len = PAGE_SIZE; + + detail->cache_request(detail, crq->item, &bp, &len); + if (len < 0) + return -EAGAIN; + return PAGE_SIZE - len; +} + static ssize_t cache_read(struct file *filp, char __user *buf, size_t count, loff_t *ppos, struct cache_detail *cd) { struct cache_reader *rp = filp->private_data; struct cache_request *rq; - struct inode *inode = filp->f_path.dentry->d_inode; + struct inode *inode = file_inode(filp); int err; if (count == 0) @@ -784,6 +796,13 @@ static ssize_t cache_read(struct file *filp, char __user *buf, size_t count, rq->readers++; spin_unlock(&queue_lock); + if (rq->len == 0) { + err = cache_request(cd, rq); + if (err < 0) + goto out; + rq->len = err; + } + if (rp->offset == 0 && !test_bit(CACHE_PENDING, &rq->item->flags)) { err = -EAGAIN; spin_lock(&queue_lock); @@ -886,7 +905,7 @@ static ssize_t cache_write(struct file *filp, const char __user *buf, struct cache_detail *cd) { struct address_space *mapping = filp->f_mapping; - struct inode *inode = filp->f_path.dentry->d_inode; + struct inode *inode = file_inode(filp); ssize_t ret = -EINVAL; if (!cd->cache_parse) @@ -1140,17 +1159,14 @@ static bool cache_listeners_exist(struct cache_detail *detail) * * Each request is at most one page long. */ -int sunrpc_cache_pipe_upcall(struct cache_detail *detail, struct cache_head *h, - void (*cache_request)(struct cache_detail *, - struct cache_head *, - char **, - int *)) +int sunrpc_cache_pipe_upcall(struct cache_detail *detail, struct cache_head *h) { char *buf; struct cache_request *crq; - char *bp; - int len; + + if (!detail->cache_request) + return -EINVAL; if (!cache_listeners_exist(detail)) { warn_no_listener(detail); @@ -1167,19 +1183,10 @@ int sunrpc_cache_pipe_upcall(struct cache_detail *detail, struct cache_head *h, return -EAGAIN; } - bp = buf; len = PAGE_SIZE; - - cache_request(detail, h, &bp, &len); - - if (len < 0) { - kfree(buf); - kfree(crq); - return -EAGAIN; - } crq->q.reader = 0; crq->item = cache_get(h); crq->buf = buf; - crq->len = PAGE_SIZE - len; + crq->len = 0; crq->readers = 0; spin_lock(&queue_lock); list_add_tail(&crq->q.list, &detail->queue); @@ -1201,7 +1208,6 @@ EXPORT_SYMBOL_GPL(sunrpc_cache_pipe_upcall); * key and content are both parsed by cache */ -#define isodigit(c) (isdigit(c) && c <= '7') int qword_get(char **bpp, char *dest, int bufsize) { /* return bytes copied, or -1 on error */ @@ -1454,7 +1460,7 @@ static ssize_t write_flush(struct file *file, const char __user *buf, static ssize_t cache_read_procfs(struct file *filp, char __user *buf, size_t count, loff_t *ppos) { - struct cache_detail *cd = PDE(filp->f_path.dentry->d_inode)->data; + struct cache_detail *cd = PDE_DATA(file_inode(filp)); return cache_read(filp, buf, count, ppos, cd); } @@ -1462,14 +1468,14 @@ static ssize_t cache_read_procfs(struct file *filp, char __user *buf, static ssize_t cache_write_procfs(struct file *filp, const char __user *buf, size_t count, loff_t *ppos) { - struct cache_detail *cd = PDE(filp->f_path.dentry->d_inode)->data; + struct cache_detail *cd = PDE_DATA(file_inode(filp)); return cache_write(filp, buf, count, ppos, cd); } static unsigned int cache_poll_procfs(struct file *filp, poll_table *wait) { - struct cache_detail *cd = PDE(filp->f_path.dentry->d_inode)->data; + struct cache_detail *cd = PDE_DATA(file_inode(filp)); return cache_poll(filp, wait, cd); } @@ -1477,22 +1483,22 @@ static unsigned int cache_poll_procfs(struct file *filp, poll_table *wait) static long cache_ioctl_procfs(struct file *filp, unsigned int cmd, unsigned long arg) { - struct inode *inode = filp->f_path.dentry->d_inode; - struct cache_detail *cd = PDE(inode)->data; + struct inode *inode = file_inode(filp); + struct cache_detail *cd = PDE_DATA(inode); return cache_ioctl(inode, filp, cmd, arg, cd); } static int cache_open_procfs(struct inode *inode, struct file *filp) { - struct cache_detail *cd = PDE(inode)->data; + struct cache_detail *cd = PDE_DATA(inode); return cache_open(inode, filp, cd); } static int cache_release_procfs(struct inode *inode, struct file *filp) { - struct cache_detail *cd = PDE(inode)->data; + struct cache_detail *cd = PDE_DATA(inode); return cache_release(inode, filp, cd); } @@ -1510,14 +1516,14 @@ static const struct file_operations cache_file_operations_procfs = { static int content_open_procfs(struct inode *inode, struct file *filp) { - struct cache_detail *cd = PDE(inode)->data; + struct cache_detail *cd = PDE_DATA(inode); return content_open(inode, filp, cd); } static int content_release_procfs(struct inode *inode, struct file *filp) { - struct cache_detail *cd = PDE(inode)->data; + struct cache_detail *cd = PDE_DATA(inode); return content_release(inode, filp, cd); } @@ -1531,14 +1537,14 @@ static const struct file_operations content_file_operations_procfs = { static int open_flush_procfs(struct inode *inode, struct file *filp) { - struct cache_detail *cd = PDE(inode)->data; + struct cache_detail *cd = PDE_DATA(inode); return open_flush(inode, filp, cd); } static int release_flush_procfs(struct inode *inode, struct file *filp) { - struct cache_detail *cd = PDE(inode)->data; + struct cache_detail *cd = PDE_DATA(inode); return release_flush(inode, filp, cd); } @@ -1546,7 +1552,7 @@ static int release_flush_procfs(struct inode *inode, struct file *filp) static ssize_t read_flush_procfs(struct file *filp, char __user *buf, size_t count, loff_t *ppos) { - struct cache_detail *cd = PDE(filp->f_path.dentry->d_inode)->data; + struct cache_detail *cd = PDE_DATA(file_inode(filp)); return read_flush(filp, buf, count, ppos, cd); } @@ -1555,7 +1561,7 @@ static ssize_t write_flush_procfs(struct file *filp, const char __user *buf, size_t count, loff_t *ppos) { - struct cache_detail *cd = PDE(filp->f_path.dentry->d_inode)->data; + struct cache_detail *cd = PDE_DATA(file_inode(filp)); return write_flush(filp, buf, count, ppos, cd); } @@ -1605,7 +1611,7 @@ static int create_cache_proc_entries(struct cache_detail *cd, struct net *net) if (p == NULL) goto out_nomem; - if (cd->cache_upcall || cd->cache_parse) { + if (cd->cache_request || cd->cache_parse) { p = proc_create_data("channel", S_IFREG|S_IRUSR|S_IWUSR, cd->u.procfs.proc_ent, &cache_file_operations_procfs, cd); @@ -1614,7 +1620,7 @@ static int create_cache_proc_entries(struct cache_detail *cd, struct net *net) goto out_nomem; } if (cd->cache_show) { - p = proc_create_data("content", S_IFREG|S_IRUSR|S_IWUSR, + p = proc_create_data("content", S_IFREG|S_IRUSR, cd->u.procfs.proc_ent, &content_file_operations_procfs, cd); cd->u.procfs.content_ent = p; @@ -1686,7 +1692,7 @@ EXPORT_SYMBOL_GPL(cache_destroy_net); static ssize_t cache_read_pipefs(struct file *filp, char __user *buf, size_t count, loff_t *ppos) { - struct cache_detail *cd = RPC_I(filp->f_path.dentry->d_inode)->private; + struct cache_detail *cd = RPC_I(file_inode(filp))->private; return cache_read(filp, buf, count, ppos, cd); } @@ -1694,14 +1700,14 @@ static ssize_t cache_read_pipefs(struct file *filp, char __user *buf, static ssize_t cache_write_pipefs(struct file *filp, const char __user *buf, size_t count, loff_t *ppos) { - struct cache_detail *cd = RPC_I(filp->f_path.dentry->d_inode)->private; + struct cache_detail *cd = RPC_I(file_inode(filp))->private; return cache_write(filp, buf, count, ppos, cd); } static unsigned int cache_poll_pipefs(struct file *filp, poll_table *wait) { - struct cache_detail *cd = RPC_I(filp->f_path.dentry->d_inode)->private; + struct cache_detail *cd = RPC_I(file_inode(filp))->private; return cache_poll(filp, wait, cd); } @@ -1709,7 +1715,7 @@ static unsigned int cache_poll_pipefs(struct file *filp, poll_table *wait) static long cache_ioctl_pipefs(struct file *filp, unsigned int cmd, unsigned long arg) { - struct inode *inode = filp->f_dentry->d_inode; + struct inode *inode = file_inode(filp); struct cache_detail *cd = RPC_I(inode)->private; return cache_ioctl(inode, filp, cmd, arg, cd); @@ -1778,7 +1784,7 @@ static int release_flush_pipefs(struct inode *inode, struct file *filp) static ssize_t read_flush_pipefs(struct file *filp, char __user *buf, size_t count, loff_t *ppos) { - struct cache_detail *cd = RPC_I(filp->f_path.dentry->d_inode)->private; + struct cache_detail *cd = RPC_I(file_inode(filp))->private; return read_flush(filp, buf, count, ppos, cd); } @@ -1787,7 +1793,7 @@ static ssize_t write_flush_pipefs(struct file *filp, const char __user *buf, size_t count, loff_t *ppos) { - struct cache_detail *cd = RPC_I(filp->f_path.dentry->d_inode)->private; + struct cache_detail *cd = RPC_I(file_inode(filp))->private; return write_flush(filp, buf, count, ppos, cd); } diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 822f020fa7f..d259fa96692 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -33,6 +33,7 @@ #include <linux/rcupdate.h> #include <linux/sunrpc/clnt.h> +#include <linux/sunrpc/addr.h> #include <linux/sunrpc/rpc_pipe_fs.h> #include <linux/sunrpc/metrics.h> #include <linux/sunrpc/bc_xprt.h> @@ -303,10 +304,8 @@ static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args, stru err = rpciod_up(); if (err) goto out_no_rpciod; - err = -EINVAL; - if (!xprt) - goto out_no_xprt; + err = -EINVAL; if (args->version >= program->nrvers) goto out_err; version = program->version[args->version]; @@ -381,10 +380,9 @@ out_no_principal: out_no_stats: kfree(clnt); out_err: - xprt_put(xprt); -out_no_xprt: rpciod_down(); out_no_rpciod: + xprt_put(xprt); return ERR_PTR(err); } @@ -413,6 +411,8 @@ struct rpc_clnt *rpc_create(struct rpc_create_args *args) }; char servername[48]; + if (args->flags & RPC_CLNT_CREATE_INFINITE_SLOTS) + xprtargs.flags |= XPRT_CREATE_INFINITE_SLOTS; /* * If the caller chooses not to specify a hostname, whip * up a string representation of the passed-in address. @@ -511,7 +511,7 @@ static struct rpc_clnt *__rpc_clone_client(struct rpc_create_args *args, new = rpc_new_client(args, xprt); if (IS_ERR(new)) { err = PTR_ERR(new); - goto out_put; + goto out_err; } atomic_inc(&clnt->cl_count); @@ -524,8 +524,6 @@ static struct rpc_clnt *__rpc_clone_client(struct rpc_create_args *args, new->cl_chatty = clnt->cl_chatty; return new; -out_put: - xprt_put(xprt); out_err: dprintk("RPC: %s: returned error %d\n", __func__, err); return ERR_PTR(err); @@ -555,7 +553,7 @@ EXPORT_SYMBOL_GPL(rpc_clone_client); * rpc_clone_client_set_auth - Clone an RPC client structure and set its auth * * @clnt: RPC client whose parameters are copied - * @auth: security flavor for new client + * @flavor: security flavor for new client * * Returns a fresh RPC client or an ERR_PTR. */ @@ -610,11 +608,6 @@ EXPORT_SYMBOL_GPL(rpc_killall_tasks); */ void rpc_shutdown_client(struct rpc_clnt *clnt) { - /* - * To avoid deadlock, never call rpc_shutdown_client from a - * workqueue context! - */ - WARN_ON_ONCE(current->flags & PF_WQ_WORKER); might_sleep(); dprintk_rcu("RPC: shutting down %s client for %s\n", @@ -1201,6 +1194,21 @@ size_t rpc_max_payload(struct rpc_clnt *clnt) EXPORT_SYMBOL_GPL(rpc_max_payload); /** + * rpc_get_timeout - Get timeout for transport in units of HZ + * @clnt: RPC client to query + */ +unsigned long rpc_get_timeout(struct rpc_clnt *clnt) +{ + unsigned long ret; + + rcu_read_lock(); + ret = rcu_dereference(clnt->cl_xprt)->timeout->to_initval; + rcu_read_unlock(); + return ret; +} +EXPORT_SYMBOL_GPL(rpc_get_timeout); + +/** * rpc_force_rebind - force transport to check that remote port is unchanged * @clnt: client to rebind * @@ -1295,6 +1303,8 @@ call_reserve(struct rpc_task *task) xprt_reserve(task); } +static void call_retry_reserve(struct rpc_task *task); + /* * 1b. Grok the result of xprt_reserve() */ @@ -1336,7 +1346,7 @@ call_reserveresult(struct rpc_task *task) case -ENOMEM: rpc_delay(task, HZ >> 2); case -EAGAIN: /* woken up; retry */ - task->tk_action = call_reserve; + task->tk_action = call_retry_reserve; return; case -EIO: /* probably a shutdown */ break; @@ -1349,6 +1359,19 @@ call_reserveresult(struct rpc_task *task) } /* + * 1c. Retry reserving an RPC call slot + */ +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); +} + +/* * 2. Bind and/or refresh the credentials */ static void @@ -1405,7 +1428,7 @@ call_allocate(struct rpc_task *task) { unsigned int slack = task->tk_rqstp->rq_cred->cr_auth->au_cslack; struct rpc_rqst *req = task->tk_rqstp; - struct rpc_xprt *xprt = task->tk_xprt; + struct rpc_xprt *xprt = req->rq_xprt; struct rpc_procinfo *proc = task->tk_msg.rpc_proc; dprint_status(task); @@ -1513,7 +1536,7 @@ rpc_xdr_encode(struct rpc_task *task) static void call_bind(struct rpc_task *task) { - struct rpc_xprt *xprt = task->tk_xprt; + struct rpc_xprt *xprt = task->tk_rqstp->rq_xprt; dprint_status(task); @@ -1607,7 +1630,7 @@ retry_timeout: static void call_connect(struct rpc_task *task) { - struct rpc_xprt *xprt = task->tk_xprt; + struct rpc_xprt *xprt = task->tk_rqstp->rq_xprt; dprintk("RPC: %5u call_connect xprt %p %s connected\n", task->tk_pid, xprt, @@ -1633,22 +1656,26 @@ call_connect_status(struct rpc_task *task) dprint_status(task); - task->tk_status = 0; - if (status >= 0 || status == -EAGAIN) { - clnt->cl_stats->netreconn++; - task->tk_action = call_transmit; - return; - } - trace_rpc_connect_status(task, status); switch (status) { /* if soft mounted, test if we've timed out */ case -ETIMEDOUT: task->tk_action = call_timeout; - break; - default: - rpc_exit(task, -EIO); + return; + case -ECONNREFUSED: + case -ECONNRESET: + case -ENETUNREACH: + if (RPC_IS_SOFTCONN(task)) + break; + /* retry with existing socket, after a delay */ + case 0: + case -EAGAIN: + task->tk_status = 0; + clnt->cl_stats->netreconn++; + task->tk_action = call_transmit; + return; } + rpc_exit(task, status); } /* @@ -1690,7 +1717,7 @@ call_transmit(struct rpc_task *task) if (rpc_reply_expected(task)) return; task->tk_action = rpc_exit_task; - rpc_wake_up_queued_task(&task->tk_xprt->pending, task); + rpc_wake_up_queued_task(&task->tk_rqstp->rq_xprt->pending, task); } /* @@ -1789,7 +1816,7 @@ call_bc_transmit(struct rpc_task *task) */ printk(KERN_NOTICE "RPC: Could not send backchannel reply " "error: %d\n", task->tk_status); - xprt_conditional_disconnect(task->tk_xprt, + xprt_conditional_disconnect(req->rq_xprt, req->rq_connect_cookie); break; default: @@ -1841,7 +1868,7 @@ call_status(struct rpc_task *task) case -ETIMEDOUT: task->tk_action = call_timeout; if (task->tk_client->cl_discrtry) - xprt_conditional_disconnect(task->tk_xprt, + xprt_conditional_disconnect(req->rq_xprt, req->rq_connect_cookie); break; case -ECONNRESET: @@ -1996,7 +2023,7 @@ out_retry: if (task->tk_rqstp == req) { req->rq_reply_bytes_recvd = req->rq_rcv_buf.len = 0; if (task->tk_client->cl_discrtry) - xprt_conditional_disconnect(task->tk_xprt, + xprt_conditional_disconnect(req->rq_xprt, req->rq_connect_cookie); } } @@ -2010,7 +2037,7 @@ rpc_encode_header(struct rpc_task *task) /* FIXME: check buffer size? */ - p = xprt_skip_transport_header(task->tk_xprt, p); + p = xprt_skip_transport_header(req->rq_xprt, p); *p++ = req->rq_xid; /* XID */ *p++ = htonl(RPC_CALL); /* CALL */ *p++ = htonl(RPC_VERSION); /* RPC version */ diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index fd10981ea79..a9129f8d707 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -284,7 +284,7 @@ out: static ssize_t rpc_pipe_read(struct file *filp, char __user *buf, size_t len, loff_t *offset) { - struct inode *inode = filp->f_path.dentry->d_inode; + struct inode *inode = file_inode(filp); struct rpc_pipe *pipe; struct rpc_pipe_msg *msg; int res = 0; @@ -328,7 +328,7 @@ out_unlock: static ssize_t rpc_pipe_write(struct file *filp, const char __user *buf, size_t len, loff_t *offset) { - struct inode *inode = filp->f_path.dentry->d_inode; + struct inode *inode = file_inode(filp); int res; mutex_lock(&inode->i_mutex); @@ -342,7 +342,7 @@ rpc_pipe_write(struct file *filp, const char __user *buf, size_t len, loff_t *of static unsigned int rpc_pipe_poll(struct file *filp, struct poll_table_struct *wait) { - struct inode *inode = filp->f_path.dentry->d_inode; + struct inode *inode = file_inode(filp); struct rpc_inode *rpci = RPC_I(inode); unsigned int mask = POLLOUT | POLLWRNORM; @@ -360,7 +360,7 @@ rpc_pipe_poll(struct file *filp, struct poll_table_struct *wait) static long rpc_pipe_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { - struct inode *inode = filp->f_path.dentry->d_inode; + struct inode *inode = file_inode(filp); struct rpc_pipe *pipe; int len; @@ -830,7 +830,7 @@ static int rpc_rmdir_depopulate(struct dentry *dentry, * responses to upcalls. They will result in calls to @msg->downcall. * * The @private argument passed here will be available to all these methods - * from the file pointer, via RPC_I(file->f_dentry->d_inode)->private. + * from the file pointer, via RPC_I(file_inode(file))->private. */ struct dentry *rpc_mkpipe_dentry(struct dentry *parent, const char *name, void *private, struct rpc_pipe *pipe) @@ -1174,6 +1174,8 @@ static struct file_system_type rpc_pipe_fs_type = { .mount = rpc_mount, .kill_sb = rpc_kill_sb, }; +MODULE_ALIAS_FS("rpc_pipefs"); +MODULE_ALIAS("rpc_pipefs"); static void init_once(void *foo) @@ -1218,6 +1220,3 @@ void unregister_rpc_pipefs(void) kmem_cache_destroy(rpc_inode_cachep); unregister_filesystem(&rpc_pipe_fs_type); } - -/* Make 'mount -t rpc_pipefs ...' autoload this module. */ -MODULE_ALIAS("rpc_pipefs"); diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c index 795a0f4e920..3df764dc330 100644 --- a/net/sunrpc/rpcb_clnt.c +++ b/net/sunrpc/rpcb_clnt.c @@ -26,6 +26,7 @@ #include <net/ipv6.h> #include <linux/sunrpc/clnt.h> +#include <linux/sunrpc/addr.h> #include <linux/sunrpc/sched.h> #include <linux/sunrpc/xprtsock.h> diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index d17a704aaf5..f8529fc8e54 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -98,9 +98,25 @@ __rpc_add_timer(struct rpc_wait_queue *queue, struct rpc_task *task) list_add(&task->u.tk_wait.timer_list, &queue->timer_list.list); } +static void rpc_rotate_queue_owner(struct rpc_wait_queue *queue) +{ + struct list_head *q = &queue->tasks[queue->priority]; + struct rpc_task *task; + + if (!list_empty(q)) { + task = list_first_entry(q, struct rpc_task, u.tk_wait.list); + if (task->tk_owner == queue->owner) + list_move_tail(&task->u.tk_wait.list, q); + } +} + static void rpc_set_waitqueue_priority(struct rpc_wait_queue *queue, int priority) { - queue->priority = priority; + if (queue->priority != priority) { + /* Fairness: rotate the list when changing priority */ + rpc_rotate_queue_owner(queue); + queue->priority = priority; + } } static void rpc_set_waitqueue_owner(struct rpc_wait_queue *queue, pid_t pid) @@ -164,6 +180,8 @@ static void __rpc_add_wait_queue(struct rpc_wait_queue *queue, list_add_tail(&task->u.tk_wait.list, &queue->tasks[0]); task->tk_waitqueue = queue; queue->qlen++; + /* 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", @@ -414,8 +432,11 @@ static void __rpc_do_wake_up_task(struct rpc_wait_queue *queue, struct rpc_task */ static void rpc_wake_up_task_queue_locked(struct rpc_wait_queue *queue, struct rpc_task *task) { - if (RPC_IS_QUEUED(task) && task->tk_waitqueue == queue) - __rpc_do_wake_up_task(queue, task); + if (RPC_IS_QUEUED(task)) { + smp_rmb(); + if (task->tk_waitqueue == queue) + __rpc_do_wake_up_task(queue, task); + } } /* @@ -934,16 +955,35 @@ struct rpc_task *rpc_new_task(const struct rpc_task_setup *setup_data) return task; } +/* + * rpc_free_task - release rpc task and perform cleanups + * + * Note that we free up the rpc_task _after_ rpc_release_calldata() + * in order to work around a workqueue dependency issue. + * + * Tejun Heo states: + * "Workqueue currently considers two work items to be the same if they're + * on the same address and won't execute them concurrently - ie. it + * makes a work item which is queued again while being executed wait + * for the previous execution to complete. + * + * If a work function frees the work item, and then waits for an event + * which should be performed by another work item and *that* work item + * recycles the freed work item, it can create a false dependency loop. + * There really is no reliable way to detect this short of verifying + * every memory free." + * + */ static void rpc_free_task(struct rpc_task *task) { - const struct rpc_call_ops *tk_ops = task->tk_ops; - void *calldata = task->tk_calldata; + unsigned short tk_flags = task->tk_flags; + + rpc_release_calldata(task->tk_ops, task->tk_calldata); - if (task->tk_flags & RPC_TASK_DYNAMIC) { + if (tk_flags & RPC_TASK_DYNAMIC) { dprintk("RPC: %5u freeing task\n", task->tk_pid); mempool_free(task, rpc_task_mempool); } - rpc_release_calldata(tk_ops, calldata); } static void rpc_async_release(struct work_struct *work) @@ -953,8 +993,7 @@ static void rpc_async_release(struct work_struct *work) static void rpc_release_resources_task(struct rpc_task *task) { - if (task->tk_rqstp) - xprt_release(task); + xprt_release(task); if (task->tk_msg.rpc_cred) { put_rpccred(task->tk_msg.rpc_cred); task->tk_msg.rpc_cred = NULL; diff --git a/net/sunrpc/stats.c b/net/sunrpc/stats.c index bc2068ee795..21b75cb08c0 100644 --- a/net/sunrpc/stats.c +++ b/net/sunrpc/stats.c @@ -64,7 +64,7 @@ static int rpc_proc_show(struct seq_file *seq, void *v) { static int rpc_proc_open(struct inode *inode, struct file *file) { - return single_open(file, rpc_proc_show, PDE(inode)->data); + return single_open(file, rpc_proc_show, PDE_DATA(inode)); } static const struct file_operations rpc_proc_fops = { diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index dbf12ac5ecb..89a588b4478 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -515,15 +515,6 @@ EXPORT_SYMBOL_GPL(svc_create_pooled); void svc_shutdown_net(struct svc_serv *serv, struct net *net) { - /* - * The set of xprts (contained in the sv_tempsocks and - * sv_permsocks lists) is now constant, since it is modified - * only by accepting new sockets (done by service threads in - * svc_recv) or aging old ones (done by sv_temptimer), or - * configuration changes (excluded by whatever locking the - * caller is using--nfsd_mutex in the case of nfsd). So it's - * safe to traverse those lists and shut everything down: - */ svc_close_net(serv, net); if (serv->sv_shutdown) @@ -1042,6 +1033,7 @@ static void svc_unregister(const struct svc_serv *serv, struct net *net) /* * dprintk the given error with the address of the client that caused it. */ +#ifdef RPC_DEBUG static __printf(2, 3) void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...) { @@ -1058,6 +1050,9 @@ void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...) va_end(args); } +#else +static __printf(2,3) void svc_printk(struct svc_rqst *rqstp, const char *fmt, ...) {} +#endif /* * Common routine for processing the RPC request. diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index b8e47fac731..80a6640f329 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -499,7 +499,8 @@ void svc_wake_up(struct svc_serv *serv) rqstp->rq_xprt = NULL; */ wake_up(&rqstp->rq_wait); - } + } else + pool->sp_task_pending = 1; spin_unlock_bh(&pool->sp_lock); } } @@ -634,7 +635,13 @@ struct svc_xprt *svc_get_next_xprt(struct svc_rqst *rqstp, long timeout) * long for cache updates. */ rqstp->rq_chandle.thread_wait = 1*HZ; + pool->sp_task_pending = 0; } else { + if (pool->sp_task_pending) { + pool->sp_task_pending = 0; + spin_unlock_bh(&pool->sp_lock); + return ERR_PTR(-EAGAIN); + } /* No data pending. Go to sleep */ svc_thread_enqueue(pool, rqstp); @@ -856,7 +863,6 @@ static void svc_age_temp_xprts(unsigned long closure) struct svc_serv *serv = (struct svc_serv *)closure; struct svc_xprt *xprt; struct list_head *le, *next; - LIST_HEAD(to_be_aged); dprintk("svc_age_temp_xprts\n"); @@ -877,25 +883,15 @@ static void svc_age_temp_xprts(unsigned long closure) if (atomic_read(&xprt->xpt_ref.refcount) > 1 || test_bit(XPT_BUSY, &xprt->xpt_flags)) continue; - svc_xprt_get(xprt); - list_move(le, &to_be_aged); + list_del_init(le); set_bit(XPT_CLOSE, &xprt->xpt_flags); set_bit(XPT_DETACHED, &xprt->xpt_flags); - } - spin_unlock_bh(&serv->sv_lock); - - while (!list_empty(&to_be_aged)) { - le = to_be_aged.next; - /* fiddling the xpt_list node is safe 'cos we're XPT_DETACHED */ - list_del_init(le); - xprt = list_entry(le, struct svc_xprt, xpt_list); - dprintk("queuing xprt %p for closing\n", xprt); /* a thread will dequeue and close it soon */ svc_xprt_enqueue(xprt); - svc_xprt_put(xprt); } + spin_unlock_bh(&serv->sv_lock); mod_timer(&serv->sv_temptimer, jiffies + svc_conn_age_period * HZ); } @@ -959,21 +955,24 @@ void svc_close_xprt(struct svc_xprt *xprt) } EXPORT_SYMBOL_GPL(svc_close_xprt); -static void svc_close_list(struct svc_serv *serv, struct list_head *xprt_list, struct net *net) +static int svc_close_list(struct svc_serv *serv, struct list_head *xprt_list, struct net *net) { struct svc_xprt *xprt; + int ret = 0; spin_lock(&serv->sv_lock); list_for_each_entry(xprt, xprt_list, xpt_list) { if (xprt->xpt_net != net) continue; + ret++; set_bit(XPT_CLOSE, &xprt->xpt_flags); - set_bit(XPT_BUSY, &xprt->xpt_flags); + svc_xprt_enqueue(xprt); } spin_unlock(&serv->sv_lock); + return ret; } -static void svc_clear_pools(struct svc_serv *serv, struct net *net) +static struct svc_xprt *svc_dequeue_net(struct svc_serv *serv, struct net *net) { struct svc_pool *pool; struct svc_xprt *xprt; @@ -988,42 +987,46 @@ static void svc_clear_pools(struct svc_serv *serv, struct net *net) if (xprt->xpt_net != net) continue; list_del_init(&xprt->xpt_ready); + spin_unlock_bh(&pool->sp_lock); + return xprt; } spin_unlock_bh(&pool->sp_lock); } + return NULL; } -static void svc_clear_list(struct svc_serv *serv, struct list_head *xprt_list, struct net *net) +static void svc_clean_up_xprts(struct svc_serv *serv, struct net *net) { struct svc_xprt *xprt; - struct svc_xprt *tmp; - LIST_HEAD(victims); - spin_lock(&serv->sv_lock); - list_for_each_entry_safe(xprt, tmp, xprt_list, xpt_list) { - if (xprt->xpt_net != net) - continue; - list_move(&xprt->xpt_list, &victims); - } - spin_unlock(&serv->sv_lock); - - list_for_each_entry_safe(xprt, tmp, &victims, xpt_list) + while ((xprt = svc_dequeue_net(serv, net))) { + set_bit(XPT_CLOSE, &xprt->xpt_flags); svc_delete_xprt(xprt); + } } +/* + * Server threads may still be running (especially in the case where the + * service is still running in other network namespaces). + * + * So we shut down sockets the same way we would on a running server, by + * setting XPT_CLOSE, enqueuing, and letting a thread pick it up to do + * the close. In the case there are no such other threads, + * threads running, svc_clean_up_xprts() does a simple version of a + * server's main event loop, and in the case where there are other + * threads, we may need to wait a little while and then check again to + * see if they're done. + */ void svc_close_net(struct svc_serv *serv, struct net *net) { - svc_close_list(serv, &serv->sv_tempsocks, net); - svc_close_list(serv, &serv->sv_permsocks, net); + int delay = 0; - svc_clear_pools(serv, net); - /* - * At this point the sp_sockets lists will stay empty, since - * svc_xprt_enqueue will not add new entries without taking the - * sp_lock and checking XPT_BUSY. - */ - svc_clear_list(serv, &serv->sv_tempsocks, net); - svc_clear_list(serv, &serv->sv_permsocks, net); + while (svc_close_list(serv, &serv->sv_permsocks, net) + + svc_close_list(serv, &serv->sv_tempsocks, net)) { + + svc_clean_up_xprts(serv, net); + msleep(delay++); + } } /* diff --git a/net/sunrpc/svcauth.c b/net/sunrpc/svcauth.c index 7963569fc04..2af7b0cba43 100644 --- a/net/sunrpc/svcauth.c +++ b/net/sunrpc/svcauth.c @@ -138,13 +138,12 @@ auth_domain_lookup(char *name, struct auth_domain *new) { struct auth_domain *hp; struct hlist_head *head; - struct hlist_node *np; head = &auth_domain_table[hash_str(name, DN_HASHBITS)]; spin_lock(&auth_domain_lock); - hlist_for_each_entry(hp, np, head, hash) { + hlist_for_each_entry(hp, head, hash) { if (strcmp(hp->name, name)==0) { kref_get(&hp->ref); spin_unlock(&auth_domain_lock); diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c index 4d012920373..c3f9e1ef7f5 100644 --- a/net/sunrpc/svcauth_unix.c +++ b/net/sunrpc/svcauth_unix.c @@ -6,6 +6,7 @@ #include <linux/sunrpc/svcsock.h> #include <linux/sunrpc/svcauth.h> #include <linux/sunrpc/gss_api.h> +#include <linux/sunrpc/addr.h> #include <linux/err.h> #include <linux/seq_file.h> #include <linux/hash.h> @@ -17,7 +18,6 @@ #include <linux/user_namespace.h> #define RPCDBG_FACILITY RPCDBG_AUTH -#include <linux/sunrpc/clnt.h> #include "netns.h" @@ -157,11 +157,6 @@ static void ip_map_request(struct cache_detail *cd, (*bpp)[-1] = '\n'; } -static int ip_map_upcall(struct cache_detail *cd, struct cache_head *h) -{ - return sunrpc_cache_pipe_upcall(cd, h, ip_map_request); -} - static struct ip_map *__ip_map_lookup(struct cache_detail *cd, char *class, struct in6_addr *addr); static int __ip_map_update(struct cache_detail *cd, struct ip_map *ipm, struct unix_domain *udom, time_t expiry); @@ -415,10 +410,15 @@ svcauth_unix_info_release(struct svc_xprt *xpt) struct unix_gid { struct cache_head h; - uid_t uid; + kuid_t uid; struct group_info *gi; }; +static int unix_gid_hash(kuid_t uid) +{ + return hash_long(from_kuid(&init_user_ns, uid), GID_HASHBITS); +} + static void unix_gid_put(struct kref *kref) { struct cache_head *item = container_of(kref, struct cache_head, ref); @@ -433,7 +433,7 @@ static int unix_gid_match(struct cache_head *corig, struct cache_head *cnew) { struct unix_gid *orig = container_of(corig, struct unix_gid, h); struct unix_gid *new = container_of(cnew, struct unix_gid, h); - return orig->uid == new->uid; + return uid_eq(orig->uid, new->uid); } static void unix_gid_init(struct cache_head *cnew, struct cache_head *citem) { @@ -465,23 +465,19 @@ static void unix_gid_request(struct cache_detail *cd, char tuid[20]; struct unix_gid *ug = container_of(h, struct unix_gid, h); - snprintf(tuid, 20, "%u", ug->uid); + snprintf(tuid, 20, "%u", from_kuid(&init_user_ns, ug->uid)); qword_add(bpp, blen, tuid); (*bpp)[-1] = '\n'; } -static int unix_gid_upcall(struct cache_detail *cd, struct cache_head *h) -{ - return sunrpc_cache_pipe_upcall(cd, h, unix_gid_request); -} - -static struct unix_gid *unix_gid_lookup(struct cache_detail *cd, uid_t uid); +static struct unix_gid *unix_gid_lookup(struct cache_detail *cd, kuid_t uid); static int unix_gid_parse(struct cache_detail *cd, char *mesg, int mlen) { /* uid expiry Ngid gid0 gid1 ... gidN-1 */ - int uid; + int id; + kuid_t uid; int gids; int rv; int i; @@ -493,9 +489,12 @@ static int unix_gid_parse(struct cache_detail *cd, return -EINVAL; mesg[mlen-1] = 0; - rv = get_int(&mesg, &uid); + rv = get_int(&mesg, &id); if (rv) return -EINVAL; + uid = make_kuid(&init_user_ns, id); + if (!uid_valid(uid)) + return -EINVAL; ug.uid = uid; expiry = get_expiry(&mesg); @@ -530,7 +529,7 @@ static int unix_gid_parse(struct cache_detail *cd, ug.h.expiry_time = expiry; ch = sunrpc_cache_update(cd, &ug.h, &ugp->h, - hash_long(uid, GID_HASHBITS)); + unix_gid_hash(uid)); if (!ch) err = -ENOMEM; else { @@ -549,7 +548,7 @@ static int unix_gid_show(struct seq_file *m, struct cache_detail *cd, struct cache_head *h) { - struct user_namespace *user_ns = current_user_ns(); + struct user_namespace *user_ns = &init_user_ns; struct unix_gid *ug; int i; int glen; @@ -565,7 +564,7 @@ static int unix_gid_show(struct seq_file *m, else glen = 0; - seq_printf(m, "%u %d:", ug->uid, glen); + seq_printf(m, "%u %d:", from_kuid_munged(user_ns, ug->uid), glen); for (i = 0; i < glen; i++) seq_printf(m, " %d", from_kgid_munged(user_ns, GROUP_AT(ug->gi, i))); seq_printf(m, "\n"); @@ -577,7 +576,7 @@ static struct cache_detail unix_gid_cache_template = { .hash_size = GID_HASHMAX, .name = "auth.unix.gid", .cache_put = unix_gid_put, - .cache_upcall = unix_gid_upcall, + .cache_request = unix_gid_request, .cache_parse = unix_gid_parse, .cache_show = unix_gid_show, .match = unix_gid_match, @@ -615,20 +614,20 @@ void unix_gid_cache_destroy(struct net *net) cache_destroy_net(cd, net); } -static struct unix_gid *unix_gid_lookup(struct cache_detail *cd, uid_t uid) +static struct unix_gid *unix_gid_lookup(struct cache_detail *cd, kuid_t uid) { struct unix_gid ug; struct cache_head *ch; ug.uid = uid; - ch = sunrpc_cache_lookup(cd, &ug.h, hash_long(uid, GID_HASHBITS)); + ch = sunrpc_cache_lookup(cd, &ug.h, unix_gid_hash(uid)); if (ch) return container_of(ch, struct unix_gid, h); else return NULL; } -static struct group_info *unix_gid_find(uid_t uid, struct svc_rqst *rqstp) +static struct group_info *unix_gid_find(kuid_t uid, struct svc_rqst *rqstp) { struct unix_gid *ug; struct group_info *gi; @@ -750,8 +749,8 @@ svcauth_null_accept(struct svc_rqst *rqstp, __be32 *authp) } /* Signal that mapping to nobody uid/gid is required */ - cred->cr_uid = (uid_t) -1; - cred->cr_gid = (gid_t) -1; + cred->cr_uid = INVALID_UID; + cred->cr_gid = INVALID_GID; cred->cr_group_info = groups_alloc(0); if (cred->cr_group_info == NULL) return SVC_CLOSE; /* kmalloc failure - client must retry */ @@ -812,8 +811,10 @@ svcauth_unix_accept(struct svc_rqst *rqstp, __be32 *authp) argv->iov_base = (void*)((__be32*)argv->iov_base + slen); /* skip machname */ argv->iov_len -= slen*4; - cred->cr_uid = svc_getnl(argv); /* uid */ - cred->cr_gid = svc_getnl(argv); /* gid */ + cred->cr_uid = make_kuid(&init_user_ns, svc_getnl(argv)); /* uid */ + cred->cr_gid = make_kgid(&init_user_ns, svc_getnl(argv)); /* gid */ + if (!uid_valid(cred->cr_uid) || !gid_valid(cred->cr_gid)) + goto badcred; slen = svc_getnl(argv); /* gids length */ if (slen > 16 || (len -= (slen + 2)*4) < 0) goto badcred; @@ -874,7 +875,7 @@ static struct cache_detail ip_map_cache_template = { .hash_size = IP_HASHMAX, .name = "auth.unix.ip", .cache_put = ip_map_put, - .cache_upcall = ip_map_upcall, + .cache_request = ip_map_request, .cache_parse = ip_map_parse, .cache_show = ip_map_show, .match = ip_map_match, diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 0a148c9d2a5..0f679df7d07 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -465,7 +465,7 @@ static int svc_udp_get_dest_address4(struct svc_rqst *rqstp, } /* - * See net/ipv6/datagram.c : datagram_recv_ctl + * See net/ipv6/datagram.c : ip6_datagram_recv_ctl */ static int svc_udp_get_dest_address6(struct svc_rqst *rqstp, struct cmsghdr *cmh) diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c index 56055632f15..75edcfad6e2 100644 --- a/net/sunrpc/xdr.c +++ b/net/sunrpc/xdr.c @@ -879,6 +879,47 @@ xdr_buf_subsegment(struct xdr_buf *buf, struct xdr_buf *subbuf, } EXPORT_SYMBOL_GPL(xdr_buf_subsegment); +/** + * xdr_buf_trim - lop at most "len" bytes off the end of "buf" + * @buf: buf to be trimmed + * @len: number of bytes to reduce "buf" by + * + * Trim an xdr_buf by the given number of bytes by fixing up the lengths. Note + * that it's possible that we'll trim less than that amount if the xdr_buf is + * too small, or if (for instance) it's all in the head and the parser has + * already read too far into it. + */ +void xdr_buf_trim(struct xdr_buf *buf, unsigned int len) +{ + size_t cur; + unsigned int trim = len; + + if (buf->tail[0].iov_len) { + cur = min_t(size_t, buf->tail[0].iov_len, trim); + buf->tail[0].iov_len -= cur; + trim -= cur; + if (!trim) + goto fix_len; + } + + if (buf->page_len) { + cur = min_t(unsigned int, buf->page_len, trim); + buf->page_len -= cur; + trim -= cur; + if (!trim) + goto fix_len; + } + + if (buf->head[0].iov_len) { + cur = min_t(size_t, buf->head[0].iov_len, trim); + buf->head[0].iov_len -= cur; + trim -= cur; + } +fix_len: + buf->len -= (len - trim); +} +EXPORT_SYMBOL_GPL(xdr_buf_trim); + static void __read_bytes_from_xdr_buf(struct xdr_buf *subbuf, void *obj, unsigned int len) { unsigned int this_len; diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index bd462a532ac..745fca3cfd3 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -430,21 +430,23 @@ __xprt_put_cong(struct rpc_xprt *xprt, struct rpc_rqst *req) */ void xprt_release_rqst_cong(struct rpc_task *task) { - __xprt_put_cong(task->tk_xprt, task->tk_rqstp); + struct rpc_rqst *req = task->tk_rqstp; + + __xprt_put_cong(req->rq_xprt, req); } EXPORT_SYMBOL_GPL(xprt_release_rqst_cong); /** * xprt_adjust_cwnd - adjust transport congestion window + * @xprt: pointer to xprt * @task: recently completed RPC request used to adjust window * @result: result code of completed RPC request * * We use a time-smoothed congestion estimator to avoid heavy oscillation. */ -void xprt_adjust_cwnd(struct rpc_task *task, int result) +void xprt_adjust_cwnd(struct rpc_xprt *xprt, struct rpc_task *task, int result) { struct rpc_rqst *req = task->tk_rqstp; - struct rpc_xprt *xprt = task->tk_xprt; unsigned long cwnd = xprt->cwnd; if (result >= 0 && cwnd <= xprt->cong) { @@ -485,13 +487,17 @@ EXPORT_SYMBOL_GPL(xprt_wake_pending_tasks); * xprt_wait_for_buffer_space - wait for transport output buffer to clear * @task: task to be put to sleep * @action: function pointer to be executed after wait + * + * Note that we only set the timer for the case of RPC_IS_SOFT(), since + * we don't in general want to force a socket disconnection due to + * an incomplete RPC call transmission. */ void xprt_wait_for_buffer_space(struct rpc_task *task, rpc_action action) { struct rpc_rqst *req = task->tk_rqstp; struct rpc_xprt *xprt = req->rq_xprt; - task->tk_timeout = req->rq_timeout; + task->tk_timeout = RPC_IS_SOFT(task) ? req->rq_timeout : 0; rpc_sleep_on(&xprt->pending, task, action); } EXPORT_SYMBOL_GPL(xprt_wait_for_buffer_space); @@ -695,7 +701,7 @@ out_abort: */ void xprt_connect(struct rpc_task *task) { - struct rpc_xprt *xprt = task->tk_xprt; + 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")); @@ -722,13 +728,13 @@ void xprt_connect(struct rpc_task *task) if (xprt_test_and_set_connecting(xprt)) return; xprt->stat.connect_start = jiffies; - xprt->ops->connect(task); + xprt->ops->connect(xprt, task); } } static void xprt_connect_status(struct rpc_task *task) { - struct rpc_xprt *xprt = task->tk_xprt; + struct rpc_xprt *xprt = task->tk_rqstp->rq_xprt; if (task->tk_status == 0) { xprt->stat.connect_count++; @@ -832,7 +838,7 @@ static void xprt_timer(struct rpc_task *task) spin_lock_bh(&xprt->transport_lock); if (!req->rq_reply_bytes_recvd) { if (xprt->ops->timer) - xprt->ops->timer(task); + xprt->ops->timer(xprt, task); } else task->tk_status = 0; spin_unlock_bh(&xprt->transport_lock); @@ -942,6 +948,34 @@ void xprt_transmit(struct rpc_task *task) spin_unlock_bh(&xprt->transport_lock); } +static void xprt_add_backlog(struct rpc_xprt *xprt, struct rpc_task *task) +{ + set_bit(XPRT_CONGESTED, &xprt->state); + rpc_sleep_on(&xprt->backlog, task, NULL); +} + +static void xprt_wake_up_backlog(struct rpc_xprt *xprt) +{ + if (rpc_wake_up_next(&xprt->backlog) == NULL) + clear_bit(XPRT_CONGESTED, &xprt->state); +} + +static bool xprt_throttle_congested(struct rpc_xprt *xprt, struct rpc_task *task) +{ + bool ret = false; + + if (!test_bit(XPRT_CONGESTED, &xprt->state)) + goto out; + spin_lock(&xprt->reserve_lock); + if (test_bit(XPRT_CONGESTED, &xprt->state)) { + rpc_sleep_on(&xprt->backlog, task, NULL); + ret = true; + } + spin_unlock(&xprt->reserve_lock); +out: + return ret; +} + static struct rpc_rqst *xprt_dynamic_alloc_slot(struct rpc_xprt *xprt, gfp_t gfp_flags) { struct rpc_rqst *req = ERR_PTR(-EAGAIN); @@ -986,7 +1020,7 @@ void xprt_alloc_slot(struct rpc_xprt *xprt, struct rpc_task *task) task->tk_status = -ENOMEM; break; case -EAGAIN: - rpc_sleep_on(&xprt->backlog, task, NULL); + xprt_add_backlog(xprt, task); dprintk("RPC: waiting for request slot\n"); default: task->tk_status = -EAGAIN; @@ -1022,7 +1056,7 @@ static void xprt_free_slot(struct rpc_xprt *xprt, struct rpc_rqst *req) memset(req, 0, sizeof(*req)); /* mark unused */ list_add(&req->rq_list, &xprt->free); } - rpc_wake_up_next(&xprt->backlog); + xprt_wake_up_backlog(xprt); spin_unlock(&xprt->reserve_lock); } @@ -1086,12 +1120,13 @@ EXPORT_SYMBOL_GPL(xprt_free); * xprt_reserve - allocate an RPC request slot * @task: RPC task requesting a slot allocation * - * If no more slots are available, place the task on the transport's + * If the transport is marked as being congested, or if no more + * slots are available, place the task on the transport's * backlog queue. */ void xprt_reserve(struct rpc_task *task) { - struct rpc_xprt *xprt = task->tk_xprt; + struct rpc_xprt *xprt; task->tk_status = 0; if (task->tk_rqstp != NULL) @@ -1099,7 +1134,36 @@ void xprt_reserve(struct rpc_task *task) task->tk_timeout = 0; task->tk_status = -EAGAIN; + rcu_read_lock(); + xprt = rcu_dereference(task->tk_client->cl_xprt); + if (!xprt_throttle_congested(xprt, task)) + xprt->ops->alloc_slot(xprt, task); + rcu_read_unlock(); +} + +/** + * xprt_retry_reserve - allocate an RPC request slot + * @task: RPC task requesting a slot allocation + * + * If no more slots are available, place the task on the transport's + * backlog queue. + * Note that the only difference with xprt_reserve is that we now + * ignore the value of the XPRT_CONGESTED flag. + */ +void xprt_retry_reserve(struct rpc_task *task) +{ + struct rpc_xprt *xprt; + + task->tk_status = 0; + if (task->tk_rqstp != NULL) + return; + + task->tk_timeout = 0; + task->tk_status = -EAGAIN; + rcu_read_lock(); + xprt = rcu_dereference(task->tk_client->cl_xprt); xprt->ops->alloc_slot(xprt, task); + rcu_read_unlock(); } static inline __be32 xprt_alloc_xid(struct rpc_xprt *xprt) @@ -1136,10 +1200,18 @@ static void xprt_request_init(struct rpc_task *task, struct rpc_xprt *xprt) void xprt_release(struct rpc_task *task) { struct rpc_xprt *xprt; - struct rpc_rqst *req; + struct rpc_rqst *req = task->tk_rqstp; - if (!(req = task->tk_rqstp)) + if (req == NULL) { + if (task->tk_client) { + rcu_read_lock(); + xprt = rcu_dereference(task->tk_client->cl_xprt); + if (xprt->snd_task == task) + xprt_release_write(xprt, task); + rcu_read_unlock(); + } return; + } xprt = req->rq_xprt; if (task->tk_ops->rpc_count_stats != NULL) diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c index 558fbab574f..e03725bfe2b 100644 --- a/net/sunrpc/xprtrdma/rpc_rdma.c +++ b/net/sunrpc/xprtrdma/rpc_rdma.c @@ -171,7 +171,7 @@ rpcrdma_create_chunks(struct rpc_rqst *rqst, struct xdr_buf *target, struct rpcrdma_msg *headerp, enum rpcrdma_chunktype type) { struct rpcrdma_req *req = rpcr_to_rdmar(rqst); - struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(rqst->rq_task->tk_xprt); + struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(rqst->rq_xprt); int nsegs, nchunks = 0; unsigned int pos; struct rpcrdma_mr_seg *seg = req->rl_segments; @@ -366,7 +366,7 @@ rpcrdma_inline_pullup(struct rpc_rqst *rqst, int pad) int rpcrdma_marshal_req(struct rpc_rqst *rqst) { - struct rpc_xprt *xprt = rqst->rq_task->tk_xprt; + struct rpc_xprt *xprt = rqst->rq_xprt; struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt); struct rpcrdma_req *req = rpcr_to_rdmar(rqst); char *base; diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c index c9aa7a35f3b..794312f22b9 100644 --- a/net/sunrpc/xprtrdma/transport.c +++ b/net/sunrpc/xprtrdma/transport.c @@ -51,6 +51,7 @@ #include <linux/init.h> #include <linux/slab.h> #include <linux/seq_file.h> +#include <linux/sunrpc/addr.h> #include "xprt_rdma.h" @@ -426,9 +427,8 @@ xprt_rdma_set_port(struct rpc_xprt *xprt, u16 port) } static void -xprt_rdma_connect(struct rpc_task *task) +xprt_rdma_connect(struct rpc_xprt *xprt, struct rpc_task *task) { - struct rpc_xprt *xprt = (struct rpc_xprt *)task->tk_xprt; struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt); if (r_xprt->rx_ep.rep_connected != 0) { @@ -475,7 +475,7 @@ xprt_rdma_reserve_xprt(struct rpc_xprt *xprt, struct rpc_task *task) static void * xprt_rdma_allocate(struct rpc_task *task, size_t size) { - struct rpc_xprt *xprt = task->tk_xprt; + struct rpc_xprt *xprt = task->tk_rqstp->rq_xprt; struct rpcrdma_req *req, *nreq; req = rpcrdma_buffer_get(&rpcx_to_rdmax(xprt)->rx_buf); @@ -627,7 +627,7 @@ static int xprt_rdma_send_request(struct rpc_task *task) { struct rpc_rqst *rqst = task->tk_rqstp; - struct rpc_xprt *xprt = task->tk_xprt; + struct rpc_xprt *xprt = rqst->rq_xprt; struct rpcrdma_req *req = rpcr_to_rdmar(rqst); struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt); diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index 745973b729a..93726560eaa 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -1086,7 +1086,7 @@ rpcrdma_buffer_create(struct rpcrdma_buffer *buf, struct rpcrdma_ep *ep, case RPCRDMA_MEMWINDOWS: /* Allocate one extra request's worth, for full cycling */ for (i = (buf->rb_max_requests+1) * RPCRDMA_MAX_SEGS; i; i--) { - r->r.mw = ib_alloc_mw(ia->ri_pd); + r->r.mw = ib_alloc_mw(ia->ri_pd, IB_MW_TYPE_1); if (IS_ERR(r->r.mw)) { rc = PTR_ERR(r->r.mw); dprintk("RPC: %s: ib_alloc_mw" @@ -1673,12 +1673,12 @@ rpcrdma_register_memwin_external(struct rpcrdma_mr_seg *seg, *nsegs = 1; rpcrdma_map_one(ia, seg, writing); - param.mr = ia->ri_bind_mem; + param.bind_info.mr = ia->ri_bind_mem; param.wr_id = 0ULL; /* no send cookie */ - param.addr = seg->mr_dma; - param.length = seg->mr_len; + param.bind_info.addr = seg->mr_dma; + param.bind_info.length = seg->mr_len; param.send_flags = 0; - param.mw_access_flags = mem_priv; + param.bind_info.mw_access_flags = mem_priv; DECR_CQCOUNT(&r_xprt->rx_ep); rc = ib_bind_mw(ia->ri_id->qp, seg->mr_chunk.rl_mw->r.mw, ¶m); @@ -1690,7 +1690,7 @@ rpcrdma_register_memwin_external(struct rpcrdma_mr_seg *seg, rpcrdma_unmap_one(ia, seg); } else { seg->mr_rkey = seg->mr_chunk.rl_mw->r.mw->rkey; - seg->mr_base = param.addr; + seg->mr_base = param.bind_info.addr; seg->mr_nsegs = 1; } return rc; @@ -1706,10 +1706,10 @@ rpcrdma_deregister_memwin_external(struct rpcrdma_mr_seg *seg, int rc; BUG_ON(seg->mr_nsegs != 1); - param.mr = ia->ri_bind_mem; - param.addr = 0ULL; /* unbind */ - param.length = 0; - param.mw_access_flags = 0; + param.bind_info.mr = ia->ri_bind_mem; + param.bind_info.addr = 0ULL; /* unbind */ + param.bind_info.length = 0; + param.bind_info.mw_access_flags = 0; if (*r) { param.wr_id = (u64) (unsigned long) *r; param.send_flags = IB_SEND_SIGNALED; diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h index 9a66c95b583..cc1445dc1d1 100644 --- a/net/sunrpc/xprtrdma/xprt_rdma.h +++ b/net/sunrpc/xprtrdma/xprt_rdma.h @@ -235,13 +235,13 @@ struct rpcrdma_create_data_internal { }; #define RPCRDMA_INLINE_READ_THRESHOLD(rq) \ - (rpcx_to_rdmad(rq->rq_task->tk_xprt).inline_rsize) + (rpcx_to_rdmad(rq->rq_xprt).inline_rsize) #define RPCRDMA_INLINE_WRITE_THRESHOLD(rq)\ - (rpcx_to_rdmad(rq->rq_task->tk_xprt).inline_wsize) + (rpcx_to_rdmad(rq->rq_xprt).inline_wsize) #define RPCRDMA_INLINE_PAD_VALUE(rq)\ - rpcx_to_rdmad(rq->rq_task->tk_xprt).padding + rpcx_to_rdmad(rq->rq_xprt).padding /* * Statistics for RPCRDMA diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 68b0a81c31d..9c2825827de 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -33,6 +33,7 @@ #include <linux/udp.h> #include <linux/tcp.h> #include <linux/sunrpc/clnt.h> +#include <linux/sunrpc/addr.h> #include <linux/sunrpc/sched.h> #include <linux/sunrpc/svcsock.h> #include <linux/sunrpc/xprtsock.h> @@ -770,7 +771,7 @@ static void xs_tcp_release_xprt(struct rpc_xprt *xprt, struct rpc_task *task) goto out_release; if (req->rq_bytes_sent == req->rq_snd_buf.len) goto out_release; - set_bit(XPRT_CLOSE_WAIT, &task->tk_xprt->state); + set_bit(XPRT_CLOSE_WAIT, &xprt->state); out_release: xprt_release_xprt(xprt, task); } @@ -848,6 +849,14 @@ static void xs_tcp_close(struct rpc_xprt *xprt) xs_tcp_shutdown(xprt); } +static void xs_local_destroy(struct rpc_xprt *xprt) +{ + xs_close(xprt); + xs_free_peer_addresses(xprt); + xprt_free(xprt); + module_put(THIS_MODULE); +} + /** * xs_destroy - prepare to shutdown a transport * @xprt: doomed transport @@ -861,10 +870,7 @@ static void xs_destroy(struct rpc_xprt *xprt) cancel_delayed_work_sync(&transport->connect_worker); - xs_close(xprt); - xs_free_peer_addresses(xprt); - xprt_free(xprt); - module_put(THIS_MODULE); + xs_local_destroy(xprt); } static inline struct rpc_xprt *xprt_from_sock(struct sock *sk) @@ -1005,7 +1011,7 @@ static void xs_udp_data_ready(struct sock *sk, int len) UDPX_INC_STATS_BH(sk, UDP_MIB_INDATAGRAMS); - xprt_adjust_cwnd(task, copied); + xprt_adjust_cwnd(xprt, task, copied); xprt_complete_rqst(task, copied); out_unlock: @@ -1646,9 +1652,9 @@ static void xs_udp_set_buffer_size(struct rpc_xprt *xprt, size_t sndsize, size_t * * Adjust the congestion window after a retransmit timeout has occurred. */ -static void xs_udp_timer(struct rpc_task *task) +static void xs_udp_timer(struct rpc_xprt *xprt, struct rpc_task *task) { - xprt_adjust_cwnd(task, -ETIMEDOUT); + xprt_adjust_cwnd(xprt, task, -ETIMEDOUT); } static unsigned short xs_get_random_port(void) @@ -1731,7 +1737,9 @@ static int xs_bind(struct sock_xprt *transport, struct socket *sock) */ static void xs_local_rpcbind(struct rpc_task *task) { - xprt_set_bound(task->tk_xprt); + rcu_read_lock(); + xprt_set_bound(rcu_dereference(task->tk_client->cl_xprt)); + rcu_read_unlock(); } static void xs_local_set_port(struct rpc_xprt *xprt, unsigned short port) @@ -1865,13 +1873,9 @@ static int xs_local_finish_connecting(struct rpc_xprt *xprt, * @xprt: RPC transport to connect * @transport: socket transport to connect * @create_sock: function to create a socket of the correct type - * - * Invoked by a work queue tasklet. */ -static void xs_local_setup_socket(struct work_struct *work) +static int xs_local_setup_socket(struct sock_xprt *transport) { - struct sock_xprt *transport = - container_of(work, struct sock_xprt, connect_worker.work); struct rpc_xprt *xprt = &transport->xprt; struct socket *sock; int status = -EIO; @@ -1916,6 +1920,30 @@ out: xprt_clear_connecting(xprt); xprt_wake_pending_tasks(xprt, status); current->flags &= ~PF_FSTRANS; + return status; +} + +static void xs_local_connect(struct rpc_xprt *xprt, struct rpc_task *task) +{ + struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); + int ret; + + if (RPC_IS_ASYNC(task)) { + /* + * We want the AF_LOCAL connect to be resolved in the + * filesystem namespace of the process making the rpc + * call. Thus we connect synchronously. + * + * If we want to support asynchronous AF_LOCAL calls, + * we'll need to figure out how to pass a namespace to + * connect. + */ + rpc_exit(task, -ENOTCONN); + return; + } + ret = xs_local_setup_socket(transport); + if (ret && !RPC_IS_SOFTCONN(task)) + msleep_interruptible(15000); } #ifdef CONFIG_SUNRPC_SWAP @@ -2179,10 +2207,6 @@ static void xs_tcp_setup_socket(struct work_struct *work) */ xs_tcp_force_close(xprt); break; - case -ECONNREFUSED: - case -ECONNRESET: - case -ENETUNREACH: - /* retry with existing socket, after a delay */ case 0: case -EINPROGRESS: case -EALREADY: @@ -2193,6 +2217,10 @@ static void xs_tcp_setup_socket(struct work_struct *work) /* Happens, for instance, if the user specified a link * local IPv6 address without a scope-id. */ + case -ECONNREFUSED: + case -ECONNRESET: + case -ENETUNREACH: + /* retry with existing socket, after a delay */ goto out; } out_eagain: @@ -2205,6 +2233,7 @@ out: /** * xs_connect - connect a socket to a remote endpoint + * @xprt: pointer to transport structure * @task: address of RPC task that manages state of connect request * * TCP: If the remote end dropped the connection, delay reconnecting. @@ -2216,9 +2245,8 @@ out: * If a UDP socket connect fails, the delay behavior here prevents * retry floods (hard mounts). */ -static void xs_connect(struct rpc_task *task) +static void xs_connect(struct rpc_xprt *xprt, struct rpc_task *task) { - struct rpc_xprt *xprt = task->tk_xprt; struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt); if (transport->sock != NULL && !RPC_IS_SOFTCONN(task)) { @@ -2453,13 +2481,13 @@ static struct rpc_xprt_ops xs_local_ops = { .alloc_slot = xprt_alloc_slot, .rpcbind = xs_local_rpcbind, .set_port = xs_local_set_port, - .connect = xs_connect, + .connect = xs_local_connect, .buf_alloc = rpc_malloc, .buf_free = rpc_free, .send_request = xs_local_send_request, .set_retrans_timeout = xprt_set_retrans_timeout_def, .close = xs_close, - .destroy = xs_destroy, + .destroy = xs_local_destroy, .print_stats = xs_local_print_stats, }; @@ -2626,8 +2654,6 @@ static struct rpc_xprt *xs_setup_local(struct xprt_create *args) goto out_err; } xprt_set_bound(xprt); - INIT_DELAYED_WORK(&transport->connect_worker, - xs_local_setup_socket); xs_format_peer_addresses(xprt, "local", RPCBIND_NETID_LOCAL); break; default: @@ -2741,9 +2767,13 @@ static struct rpc_xprt *xs_setup_tcp(struct xprt_create *args) struct rpc_xprt *xprt; struct sock_xprt *transport; struct rpc_xprt *ret; + unsigned int max_slot_table_size = xprt_max_tcp_slot_table_entries; + + if (args->flags & XPRT_CREATE_INFINITE_SLOTS) + max_slot_table_size = RPC_MAX_SLOT_TABLE_LIMIT; xprt = xs_setup_xprt(args, xprt_tcp_slot_table_entries, - xprt_max_tcp_slot_table_entries); + max_slot_table_size); if (IS_ERR(xprt)) return xprt; transport = container_of(xprt, struct sock_xprt, xprt); |