Viewing: mdt_lib.c
// SPDX-License-Identifier: GPL-2.0
/*
* Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved.
* Use is subject to license terms.
*
* Copyright (c) 2011, 2017, Intel Corporation.
*/
/*
* This file is part of Lustre, http://www.lustre.org/
*
* Lustre Metadata Target (mdt) request unpacking helper.
*
* Author: Peter Braam <braam@clusterfs.com>
* Author: Andreas Dilger <adilger@clusterfs.com>
* Author: Phil Schwan <phil@clusterfs.com>
* Author: Mike Shaver <shaver@clusterfs.com>
* Author: Nikita Danilov <nikita@clusterfs.com>
* Author: Huang Hua <huanghua@clusterfs.com>
* Author: Fan Yong <fanyong@clusterfs.com>
*/
#define DEBUG_SUBSYSTEM S_MDS
#include <linux/user_namespace.h>
#include <linux/uidgid.h>
#include "mdt_internal.h"
#include <uapi/linux/lnet/nidstr.h>
#include <lustre_compat.h>
#include <lustre_nodemap.h>
typedef enum ucred_init_type {
NONE_INIT = 0,
BODY_INIT = 1,
REC_INIT = 2
} ucred_init_type_t;
static __u64 get_mrc_cr_flags(struct mdt_rec_create *mrc)
{
return (__u64)(mrc->cr_flags_l) | ((__u64)mrc->cr_flags_h << 32);
}
static struct upcall_cache *get_cache(struct mdt_thread_info *info)
{
/* If nodemap does not have server_upcall role,
* use dedicated INTERNAL upcall cache, unless
* Kerberos is enforced, in which case we do not trust the client's
* provided supplementary groups.
*/
struct ptlrpc_request *req = mdt_info_req(info);
if (SPTLRPC_FLVR_MECH(req->rq_flvr.sf_rpc) == SPTLRPC_MECH_GSS_KRB5 ||
mdt_ucred(info)->uc_rbac_server_upcall)
return info->mti_mdt->mdt_identity_cache;
return info->mti_mdt->mdt_identity_cache_int;
}
void mdt_exit_ucred(struct mdt_thread_info *info)
{
struct lu_ucred *uc = mdt_ucred(info);
LASSERT(uc != NULL);
if (uc->uc_valid != UCRED_INIT) {
uc->uc_suppgids[0] = uc->uc_suppgids[1] = -1;
if (uc->uc_ginfo) {
put_group_info(uc->uc_ginfo);
uc->uc_ginfo = NULL;
}
if (uc->uc_identity) {
mdt_identity_put(get_cache(info), uc->uc_identity);
uc->uc_identity = NULL;
}
uc->uc_valid = UCRED_INIT;
}
}
static int match_nosquash_list(struct spinlock *rsi_lock,
struct list_head *nidlist,
struct lnet_nid *peernid)
{
int rc;
ENTRY;
spin_lock(rsi_lock);
rc = cfs_match_nid(peernid, nidlist);
spin_unlock(rsi_lock);
RETURN(rc);
}
/* root_squash for inter-MDS operations */
static int mdt_root_squash(struct mdt_thread_info *info,
struct lnet_nid *peernid)
{
struct lu_ucred *ucred = mdt_ucred(info);
struct root_squash_info *squash = &info->mti_mdt->mdt_squash;
ENTRY;
LASSERT(ucred != NULL);
if (!squash->rsi_uid || ucred->uc_fsuid)
RETURN(0);
if (match_nosquash_list(&squash->rsi_lock,
&squash->rsi_nosquash_nids,
peernid)) {
CDEBUG(D_OTHER, "%s is in nosquash_nids list\n",
libcfs_nidstr(peernid));
RETURN(0);
}
CDEBUG(D_OTHER, "squash req from %s, (%d:%d/%x)=>(%d:%d/%x)\n",
libcfs_nidstr(peernid), ucred->uc_fsuid, ucred->uc_fsgid,
(u32)ll_capability_u32(ucred->uc_cap),
squash->rsi_uid, squash->rsi_gid, 0);
ucred->uc_fsuid = squash->rsi_uid;
ucred->uc_fsgid = squash->rsi_gid;
ucred->uc_cap = CAP_EMPTY_SET;
ucred->uc_suppgids[0] = -1;
ucred->uc_suppgids[1] = -1;
if (ucred->uc_ginfo) {
put_group_info(ucred->uc_ginfo);
ucred->uc_ginfo = NULL;
}
RETURN(0);
}
static void ucred_set_jobid(struct mdt_thread_info *info, struct lu_ucred *uc)
{
struct ptlrpc_request *req = mdt_info_req(info);
const char *jobid = mdt_req_get_jobid(req);
/* set jobid if specified. */
if (jobid)
strscpy(uc->uc_jobid, jobid, sizeof(uc->uc_jobid));
else
uc->uc_jobid[0] = '\0';
}
static void ucred_set_nid(struct mdt_thread_info *info, struct lu_ucred *uc)
{
if (info && info->mti_exp && info->mti_exp->exp_connection)
uc->uc_nid = info->mti_exp->exp_connection->c_peer.nid;
else
uc->uc_nid = LNET_ANY_NID;
}
static void ucred_set_audit_enabled(struct mdt_thread_info *info,
struct lu_ucred *uc)
{
struct lu_nodemap *nodemap = NULL;
bool audit = true;
if (info && info->mti_exp) {
nodemap = nodemap_get_from_exp(info->mti_exp);
if (nodemap && !IS_ERR(nodemap)) {
audit = nodemap->nmf_enable_audit;
nodemap_putref(nodemap);
}
}
uc->uc_enable_audit = audit;
}
static void ucred_set_rbac_roles(struct mdt_thread_info *info,
struct lu_ucred *uc)
{
struct lu_nodemap *nodemap = NULL;
enum nodemap_rbac_roles rbac = NODEMAP_RBAC_ALL;
if (info && info->mti_exp) {
nodemap = nodemap_get_from_exp(info->mti_exp);
if (!IS_ERR_OR_NULL(nodemap)) {
rbac = nodemap->nmf_rbac;
nodemap_putref(nodemap);
}
}
uc->uc_rbac_file_perms = !!(rbac & NODEMAP_RBAC_FILE_PERMS);
uc->uc_rbac_dne_ops = !!(rbac & NODEMAP_RBAC_DNE_OPS);
uc->uc_rbac_quota_ops = !!(rbac & NODEMAP_RBAC_QUOTA_OPS);
uc->uc_rbac_byfid_ops = !!(rbac & NODEMAP_RBAC_BYFID_OPS);
uc->uc_rbac_chlg_ops = !!(rbac & NODEMAP_RBAC_CHLG_OPS);
uc->uc_rbac_fscrypt_admin = !!(rbac & NODEMAP_RBAC_FSCRYPT_ADMIN);
uc->uc_rbac_server_upcall = !!(rbac & NODEMAP_RBAC_SERVER_UPCALL);
uc->uc_rbac_ignore_root_prjquota =
!!(rbac & NODEMAP_RBAC_IGN_ROOT_PRJQUOTA);
uc->uc_rbac_hsm_ops = !!(rbac & NODEMAP_RBAC_HSM_OPS);
uc->uc_rbac_local_admin = !!(rbac & NODEMAP_RBAC_LOCAL_ADMIN);
uc->uc_rbac_pool_quota_ops = !!(rbac & NODEMAP_RBAC_POOL_QUOTA_OPS);
uc->uc_rbac_lqa_quota_ops = !!(rbac & NODEMAP_RBAC_LQA_QUOTA_OPS);
uc->uc_rbac_projid_set = !!(rbac & NODEMAP_RBAC_PROJID_SET);
uc->uc_rbac_foreign_ops = !!(rbac & NODEMAP_RBAC_FOREIGN_OPS);
uc->uc_rbac_immutable_flags = !!(rbac & NODEMAP_RBAC_IMMUTABLE_FLAGS);
}
static int new_init_ucred(struct mdt_thread_info *info, ucred_init_type_t type,
void *buf)
{
struct ptlrpc_request *req = mdt_info_req(info);
struct mdt_device *mdt = info->mti_mdt;
struct ptlrpc_user_desc *pud = req->rq_user_desc;
struct lu_ucred *ucred = mdt_ucred(info);
struct lu_nodemap *nodemap;
struct lnet_nid peernid = req->rq_peer.nid;
struct md_identity *identity = NULL;
__u32 perm = 0;
int setuid;
int setgid;
int i, rc = 0;
ENTRY;
LASSERT(req->rq_auth_gss);
LASSERT(!req->rq_auth_usr_mdt);
LASSERT(req->rq_user_desc);
LASSERT(ucred != NULL);
ucred->uc_valid = UCRED_INVALID;
ucred->uc_o_uid = pud->pud_uid;
ucred->uc_o_gid = pud->pud_gid;
ucred->uc_o_fsuid = pud->pud_fsuid;
ucred->uc_o_fsgid = pud->pud_fsgid;
nodemap = nodemap_get_from_exp(info->mti_exp);
if (IS_ERR(nodemap))
RETURN(PTR_ERR(nodemap));
pud->pud_uid = nodemap_map_id(nodemap, NODEMAP_UID,
NODEMAP_CLIENT_TO_FS, pud->pud_uid);
pud->pud_gid = nodemap_map_id(nodemap, NODEMAP_GID,
NODEMAP_CLIENT_TO_FS, pud->pud_gid);
pud->pud_fsuid = nodemap_map_id(nodemap, NODEMAP_UID,
NODEMAP_CLIENT_TO_FS, pud->pud_fsuid);
pud->pud_fsgid = nodemap_map_id(nodemap, NODEMAP_GID,
NODEMAP_CLIENT_TO_FS, pud->pud_fsgid);
for (i = 0; i < pud->pud_ngroups; i++)
pud->pud_groups[i] = nodemap_map_suppgid(nodemap,
pud->pud_groups[i]);
ucred->uc_uid = pud->pud_uid;
ucred->uc_gid = pud->pud_gid;
ucred->uc_fsuid = pud->pud_fsuid;
ucred->uc_fsgid = pud->pud_fsgid;
ucred->uc_cap = CAP_EMPTY_SET;
ll_set_capability_u32(&ucred->uc_cap, pud->pud_cap);
if (type == BODY_INIT) {
struct mdt_body *body = (struct mdt_body *)buf;
ucred->uc_suppgids[0] = nodemap_map_suppgid(nodemap,
body->mbo_suppgid);
ucred->uc_suppgids[1] = -1;
}
/* Perm checks before fetching external identity */
if (!flvr_is_rootonly(req->rq_flvr.sf_rpc) &&
req->rq_auth_uid != pud->pud_uid) {
CDEBUG(D_SEC, "local client %s: auth uid %u while client claims %u:%u/%u:%u\n",
libcfs_nidstr(&peernid), req->rq_auth_uid,
pud->pud_uid, pud->pud_gid,
pud->pud_fsuid, pud->pud_fsgid);
GOTO(out_nodemap, rc = -EACCES);
}
if (nodemap && nodemap->nmf_deny_unknown &&
nodemap_id_is_squashed(nodemap, ucred->uc_o_uid, NODEMAP_UID,
NODEMAP_CLIENT_TO_FS))
/* deny access before we get identity ref */
GOTO(out, rc = -EACCES);
ucred_set_rbac_roles(info, ucred);
/* Fetch external identity info, if enabled */
if (!is_identity_get_disabled(get_cache(info))) {
identity = mdt_identity_get(get_cache(info),
pud->pud_uid, info);
if (IS_ERR(identity)) {
if (unlikely(PTR_ERR(identity) == -EREMCHG ||
cap_raised(ucred->uc_cap,
CAP_DAC_READ_SEARCH))) {
identity = NULL;
} else {
CDEBUG(D_SEC,
"Deny access without identity: uid %u\n",
pud->pud_uid);
GOTO(out_nodemap, rc = -EACCES);
}
}
}
ucred->uc_identity = identity;
/* Perm checks that needs external identity */
if (ucred->uc_identity)
perm = mdt_identity_get_perm(ucred->uc_identity, &peernid);
else
perm = CFS_SETUID_PERM | CFS_SETGID_PERM | CFS_SETGRP_PERM;
/* find out the setuid/setgid attempt */
setuid = (pud->pud_uid != pud->pud_fsuid);
setgid = ((pud->pud_gid != pud->pud_fsgid) ||
(ucred->uc_identity &&
(pud->pud_gid != ucred->uc_identity->mi_gid)));
/* check permission of setuid */
if (setuid && !(perm & CFS_SETUID_PERM)) {
CDEBUG(D_SEC, "mdt blocked setuid attempt (%u -> %u) from %s\n",
pud->pud_uid, pud->pud_fsuid, libcfs_nidstr(&peernid));
GOTO(out, rc = -EACCES);
}
/* check permission of setgid */
if (setgid && !(perm & CFS_SETGID_PERM)) {
CDEBUG(D_SEC,
"mdt blocked setgid attempt (%u:%u/%u:%u -> %d) from %s\n",
pud->pud_uid, pud->pud_gid,
pud->pud_fsuid, pud->pud_fsgid,
ucred->uc_identity ? ucred->uc_identity->mi_gid : -1,
libcfs_nidstr(&peernid));
GOTO(out, rc = -EACCES);
}
if (perm & CFS_SETGRP_PERM && pud->pud_ngroups) {
/* setgroups for local client */
ucred->uc_ginfo = groups_alloc(pud->pud_ngroups);
if (!ucred->uc_ginfo) {
CERROR("failed to alloc %d groups\n",
pud->pud_ngroups);
GOTO(out, rc = -ENOMEM);
}
lustre_groups_from_list(ucred->uc_ginfo, pud->pud_groups);
lustre_groups_sort(ucred->uc_ginfo);
} else {
ucred->uc_suppgids[0] = -1;
ucred->uc_suppgids[1] = -1;
ucred->uc_ginfo = NULL;
}
/* clear suppgids if uid or gid was squashed. */
if (nodemap_id_is_squashed(nodemap, ucred->uc_o_uid, NODEMAP_UID,
NODEMAP_CLIENT_TO_FS) ||
nodemap_id_is_squashed(nodemap, ucred->uc_o_gid, NODEMAP_GID,
NODEMAP_CLIENT_TO_FS)) {
ucred->uc_cap = CAP_EMPTY_SET;
ucred->uc_suppgids[0] = -1;
ucred->uc_suppgids[1] = -1;
if (ucred->uc_ginfo) {
put_group_info(ucred->uc_ginfo);
ucred->uc_ginfo = NULL;
}
}
mdt_root_squash(info, &peernid);
if (!is_local_root(ucred->uc_fsuid, nodemap)) {
kernel_cap_t caps;
bool from_nodemap;
from_nodemap = nodemap &&
nodemap->nmf_caps_type != NODEMAP_CAP_OFF;
caps = from_nodemap ? nodemap->nm_capabilities :
mdt->mdt_enable_cap_mask;
if (!cap_issubset(ucred->uc_cap, caps))
CDEBUG(D_SEC, "%s: drop capabilities %llx for NID %s\n",
mdt_obd_name(mdt),
#ifdef CAP_FOR_EACH_U32
ucred->uc_cap.cap[0] |
((u64)ucred->uc_cap.cap[1] << 32),
#else
ucred->uc_cap.val,
#endif
libcfs_nidstr(&mdt_info_req(info)->rq_peer.nid));
if (!from_nodemap || nodemap->nmf_caps_type == NODEMAP_CAP_MASK)
ucred->uc_cap = cap_intersect(ucred->uc_cap, caps);
else
ucred->uc_cap = caps;
}
/* Thanks to Kerberos, Lustre does not have to trust clients anymore,
* but relies on keytabs and tickets, cryptographically validated, to
* recognize clients and users.
* RPC provided primary group should match the one got from the
* identity upcall.
* And RPC provided supplementary groups should not be trusted,
* but checked thanks to identity upcall and the trusted UID
* from the ticket.
*/
if (SPTLRPC_FLVR_MECH(req->rq_flvr.sf_rpc) == SPTLRPC_MECH_GSS_KRB5) {
if (!has_proper_groups(ucred))
GOTO(out, rc = -EACCES);
ucred->uc_suppgids[0] = -1;
ucred->uc_suppgids[1] = -1;
if (ucred->uc_ginfo) {
put_group_info(ucred->uc_ginfo);
ucred->uc_ginfo = NULL;
}
}
ucred_set_jobid(info, ucred);
ucred_set_nid(info, ucred);
ucred_set_audit_enabled(info, ucred);
ucred->uc_valid = UCRED_NEW;
EXIT;
out:
if (rc) {
if (ucred->uc_ginfo) {
put_group_info(ucred->uc_ginfo);
ucred->uc_ginfo = NULL;
}
if (ucred->uc_identity) {
mdt_identity_put(get_cache(info), ucred->uc_identity);
ucred->uc_identity = NULL;
}
}
out_nodemap:
nodemap_putref(nodemap);
return rc;
}
/**
* allow_client_chgrp() - Check whether allow the client to set supplementary
* group IDs or not.
* @info: pointer to the thread context
* @uc: pointer to the RPC user descriptor
*
* Return:
* * %true if allow to set supplementary group IDs
* * %false for other cases
*/
bool allow_client_chgrp(struct mdt_thread_info *info, struct lu_ucred *uc)
{
__u32 perm;
/* 1. identity_upcall disabled? permit local client to do anything. */
if (is_identity_get_disabled(info->mti_mdt->mdt_identity_cache))
return true;
/* 2. If fail to get related identities, then forbid any client to
* set supplementary group IDs.
*/
if (uc->uc_identity == NULL)
return false;
/* 3. Check the permission in the identities. */
perm = mdt_identity_get_perm(
uc->uc_identity,
&mdt_info_req(info)->rq_peer.nid);
if (perm & CFS_SETGRP_PERM)
return true;
return false;
}
int mdt_check_ucred(struct mdt_thread_info *info)
{
struct ptlrpc_request *req = mdt_info_req(info);
struct mdt_device *mdt = info->mti_mdt;
struct ptlrpc_user_desc *pud = req->rq_user_desc;
struct lu_ucred *ucred = mdt_ucred(info);
struct md_identity *identity = NULL;
struct lnet_nid peernid = req->rq_peer.nid;
__u32 perm = 0;
int setuid;
int setgid;
int rc = 0;
ENTRY;
LASSERT(ucred != NULL);
if ((ucred->uc_valid == UCRED_OLD) || (ucred->uc_valid == UCRED_NEW))
RETURN(0);
if (!req->rq_auth_gss || req->rq_auth_usr_mdt || !req->rq_user_desc)
RETURN(0);
/* sanity check: if we use strong authentication, we expect the
* uid which client claimed is true
*/
if (!flvr_is_rootonly(req->rq_flvr.sf_rpc) &&
req->rq_auth_uid != pud->pud_uid) {
CDEBUG(D_SEC, "local client %s: auth uid %u while client claims %u:%u/%u:%u\n",
libcfs_nidstr(&peernid), req->rq_auth_uid,
pud->pud_uid, pud->pud_gid,
pud->pud_fsuid, pud->pud_fsgid);
RETURN(-EACCES);
}
if (is_identity_get_disabled(mdt->mdt_identity_cache))
RETURN(0);
identity = mdt_identity_get(mdt->mdt_identity_cache, pud->pud_uid,
info);
if (IS_ERR(identity)) {
if (unlikely(PTR_ERR(identity) == -EREMCHG)) {
RETURN(0);
} else {
CDEBUG(D_SEC, "Deny access without identity: uid %u\n",
pud->pud_uid);
RETURN(-EACCES);
}
}
perm = mdt_identity_get_perm(identity, &peernid);
/* find out the setuid/setgid attempt */
setuid = (pud->pud_uid != pud->pud_fsuid);
setgid = (pud->pud_gid != pud->pud_fsgid ||
pud->pud_gid != identity->mi_gid);
/* check permission of setuid */
if (setuid && !(perm & CFS_SETUID_PERM)) {
CDEBUG(D_SEC, "mdt blocked setuid attempt (%u -> %u) from %s\n",
pud->pud_uid, pud->pud_fsuid, libcfs_nidstr(&peernid));
GOTO(out, rc = -EACCES);
}
/* check permission of setgid */
if (setgid && !(perm & CFS_SETGID_PERM)) {
CDEBUG(D_SEC,
"mdt blocked setgid attempt (%u:%u/%u:%u -> %u) from %s\n",
pud->pud_uid, pud->pud_gid,
pud->pud_fsuid, pud->pud_fsgid, identity->mi_gid,
libcfs_nidstr(&peernid));
GOTO(out, rc = -EACCES);
}
EXIT;
out:
mdt_identity_put(mdt->mdt_identity_cache, identity);
return rc;
}
static int old_init_ucred_common(struct mdt_thread_info *info,
struct lu_nodemap *nodemap)
{
struct ptlrpc_request *req = mdt_info_req(info);
struct lu_ucred *uc = mdt_ucred(info);
struct mdt_device *mdt = info->mti_mdt;
struct md_identity *identity = NULL;
if (nodemap && nodemap->nmf_deny_unknown &&
nodemap_id_is_squashed(nodemap, uc->uc_o_uid, NODEMAP_UID,
NODEMAP_CLIENT_TO_FS))
/* deny access before we get identity ref */
RETURN(-EACCES);
ucred_set_rbac_roles(info, uc);
if (!is_identity_get_disabled(get_cache(info))) {
identity = mdt_identity_get(get_cache(info),
uc->uc_fsuid, info);
if (IS_ERR(identity)) {
if (unlikely(PTR_ERR(identity) == -EREMCHG ||
cap_raised(uc->uc_cap,
CAP_DAC_READ_SEARCH))) {
identity = NULL;
} else {
CDEBUG(D_SEC, "Deny access without identity: uid %u\n",
uc->uc_fsuid);
RETURN(-EACCES);
}
}
}
uc->uc_identity = identity;
if (nodemap_id_is_squashed(nodemap, uc->uc_o_uid, NODEMAP_UID,
NODEMAP_CLIENT_TO_FS) ||
nodemap_id_is_squashed(nodemap, uc->uc_o_gid, NODEMAP_GID,
NODEMAP_CLIENT_TO_FS)) {
uc->uc_cap = CAP_EMPTY_SET;
uc->uc_suppgids[0] = -1;
uc->uc_suppgids[1] = -1;
}
/* process root_squash here. */
mdt_root_squash(info,
&mdt_info_req(info)->rq_peer.nid);
if (!is_local_root(uc->uc_fsuid, nodemap)) {
kernel_cap_t caps;
bool from_nodemap;
from_nodemap = nodemap &&
nodemap->nmf_caps_type != NODEMAP_CAP_OFF;
caps = from_nodemap ? nodemap->nm_capabilities :
mdt->mdt_enable_cap_mask;
if (!cap_issubset(uc->uc_cap, caps))
CDEBUG(D_SEC, "%s: drop capabilities %llx for NID %s\n",
mdt_obd_name(mdt),
#ifdef CAP_FOR_EACH_U32
uc->uc_cap.cap[0] | ((u64)uc->uc_cap.cap[1]<<32),
#else
uc->uc_cap.val,
#endif
libcfs_nidstr(&mdt_info_req(info)->rq_peer.nid));
if (!from_nodemap || nodemap->nmf_caps_type == NODEMAP_CAP_MASK)
uc->uc_cap = cap_intersect(uc->uc_cap, caps);
else
uc->uc_cap = caps;
}
/* Thanks to Kerberos, Lustre does not have to trust clients anymore,
* but relies on keytabs and tickets, cryptographically validated, to
* recognize clients and users.
* RPC provided primary group should match the one got from the
* identity upcall.
* And RPC provided supplementary groups should not be trusted,
* but checked thanks to identity upcall and the trusted UID
* from the ticket.
*/
if (SPTLRPC_FLVR_MECH(req->rq_flvr.sf_rpc) == SPTLRPC_MECH_GSS_KRB5) {
if (!has_proper_groups(uc)) {
mdt_identity_put(mdt->mdt_identity_cache,
uc->uc_identity);
uc->uc_identity = NULL;
RETURN(-EACCES);
}
uc->uc_suppgids[0] = -1;
uc->uc_suppgids[1] = -1;
}
ucred_set_jobid(info, uc);
ucred_set_nid(info, uc);
ucred_set_audit_enabled(info, uc);
uc->uc_valid = UCRED_OLD;
RETURN(0);
}
static int old_init_ucred(struct mdt_thread_info *info,
struct mdt_body *body)
{
struct lu_ucred *uc = mdt_ucred(info);
struct lu_nodemap *nodemap;
int rc;
ENTRY;
uc->uc_o_uid = body->mbo_uid;
uc->uc_o_gid = body->mbo_gid;
uc->uc_o_fsuid = body->mbo_fsuid;
uc->uc_o_fsgid = body->mbo_fsgid;
nodemap = nodemap_get_from_exp(info->mti_exp);
if (IS_ERR(nodemap))
RETURN(PTR_ERR(nodemap));
body->mbo_uid = nodemap_map_id(nodemap, NODEMAP_UID,
NODEMAP_CLIENT_TO_FS, body->mbo_uid);
body->mbo_gid = nodemap_map_id(nodemap, NODEMAP_GID,
NODEMAP_CLIENT_TO_FS, body->mbo_gid);
body->mbo_fsuid = nodemap_map_id(nodemap, NODEMAP_UID,
NODEMAP_CLIENT_TO_FS, body->mbo_fsuid);
body->mbo_fsgid = nodemap_map_id(nodemap, NODEMAP_GID,
NODEMAP_CLIENT_TO_FS, body->mbo_fsgid);
LASSERT(uc != NULL);
uc->uc_valid = UCRED_INVALID;
uc->uc_uid = body->mbo_uid;
uc->uc_gid = body->mbo_gid;
uc->uc_fsuid = body->mbo_fsuid;
uc->uc_fsgid = body->mbo_fsgid;
uc->uc_suppgids[0] = nodemap_map_suppgid(nodemap, body->mbo_suppgid);
uc->uc_suppgids[1] = -1;
uc->uc_ginfo = NULL;
uc->uc_cap = CAP_EMPTY_SET;
ll_set_capability_u32(&uc->uc_cap, body->mbo_capability);
rc = old_init_ucred_common(info, nodemap);
nodemap_putref(nodemap);
RETURN(rc);
}
static int old_init_ucred_reint(struct mdt_thread_info *info)
{
struct lu_ucred *uc = mdt_ucred(info);
struct lu_nodemap *nodemap;
int rc;
ENTRY;
uc->uc_o_uid = uc->uc_o_fsuid = uc->uc_fsuid;
uc->uc_o_gid = uc->uc_o_fsgid = uc->uc_fsgid;
nodemap = nodemap_get_from_exp(info->mti_exp);
if (IS_ERR(nodemap))
RETURN(PTR_ERR(nodemap));
LASSERT(uc != NULL);
uc->uc_fsuid = nodemap_map_id(nodemap, NODEMAP_UID,
NODEMAP_CLIENT_TO_FS, uc->uc_fsuid);
uc->uc_fsgid = nodemap_map_id(nodemap, NODEMAP_GID,
NODEMAP_CLIENT_TO_FS, uc->uc_fsgid);
uc->uc_valid = UCRED_INVALID;
uc->uc_uid = uc->uc_fsuid;
uc->uc_gid = uc->uc_fsgid;
uc->uc_ginfo = NULL;
rc = old_init_ucred_common(info, nodemap);
nodemap_putref(nodemap);
RETURN(rc);
}
int mdt_init_ucred(struct mdt_thread_info *info, struct mdt_body *body)
{
struct ptlrpc_request *req = mdt_info_req(info);
struct lu_ucred *uc = mdt_ucred(info);
LASSERT(uc != NULL);
if ((uc->uc_valid == UCRED_OLD) || (uc->uc_valid == UCRED_NEW))
return 0;
mdt_exit_ucred(info);
if (!req->rq_auth_gss || req->rq_auth_usr_mdt || !req->rq_user_desc)
return old_init_ucred(info, body);
else
return new_init_ucred(info, BODY_INIT, body);
}
int mdt_init_ucred_reint(struct mdt_thread_info *info)
{
struct ptlrpc_request *req = mdt_info_req(info);
struct lu_ucred *uc = mdt_ucred(info);
struct md_attr *ma = &info->mti_attr;
LASSERT(uc != NULL);
if ((uc->uc_valid == UCRED_OLD) || (uc->uc_valid == UCRED_NEW))
return 0;
/* LU-5564: for normal close request, skip permission check */
if (lustre_msg_get_opc(req->rq_reqmsg) == MDS_CLOSE &&
!(ma->ma_attr_flags & (MDS_HSM_RELEASE | MDS_CLOSE_LAYOUT_SWAP))) {
cap_raise_nfsd_set(uc->uc_cap, CAP_FULL_SET);
cap_raise_fs_set(uc->uc_cap, CAP_FULL_SET);
}
mdt_exit_ucred(info);
if (!req->rq_auth_gss || req->rq_auth_usr_mdt || !req->rq_user_desc)
return old_init_ucred_reint(info);
else
return new_init_ucred(info, REC_INIT, NULL);
}
/**
* mdt_enable_gid_deny - check if operation allowed by enable_XXX_gid
*
* @uc: process credentials
* @cap: capability required to override permission checks
* @ena_gid: enable_foo_gid value that specifies required GID
*
* Returns false if request allowed to perform action, true if request is denied
*/
bool mdt_enable_gid_deny(struct lu_ucred *uc, unsigned int cap, gid_t ena_gid)
{
if (unlikely(!uc)) /* cannot check permissions */
RETURN(false);
CDEBUG(D_INFO,
"cap=%x raised=%u ena_gid=%u gid=%u fsgid=%u ingrp=%u\n", cap,
cap_raised(uc->uc_cap, cap), ena_gid, uc->uc_gid, uc->uc_fsgid,
lustre_in_group_p(uc, ena_gid));
if (ena_gid == MDT_INVALID_GID) /* any GID is allowed */
GOTO(out_allow, false);
if (cap_raised(uc->uc_cap, cap)) /* process has capability set */
GOTO(out_allow, false);
if (lustre_in_group_p(uc, ena_gid)) /* process has required GID */
GOTO(out_allow, false);
lustre_print_ucred(uc);
RETURN(true); /* process missing needed GID */
out_allow:
return false;
}
/**
* mdt_check_resource_ids() - check client access to resource via nodemap
*
* @info: mdt thread environment
* @obj: mdt object to check
*
* Check whether the client is allowed to access the resource by consulting
* the nodemap with the client's export and the MDT inode's UID/GID attributes.
*
* Return:
* * %0 on success (access is allowed)
* * %-ECHRNG if access is denied
*/
int mdt_check_resource_ids(struct mdt_thread_info *info, struct mdt_object *obj)
{
struct dt_object *dt;
struct lu_attr la = { 0 };
ENTRY;
if (info->mti_mdt->mdt_lut.lut_enable_resource_id_check == 0)
RETURN(0);
dt = mdt_obj2dt(obj);
/* Get attributes from MDT inode */
if (dt && dt->do_ops && dt->do_ops->do_attr_get) {
dt_attr_get(info->mti_env, mdt_obj2dt(obj), &la);
} else {
/* log this case but don't return err code */
CERROR("%s: no dt object for " DFID ": rc = %d\n",
mdt_obd_name(info->mti_mdt), PFID(mdt_object_fid(obj)),
-ENOENT);
RETURN(0);
}
RETURN(nodemap_check_resource_ids(mdt_info_req(info)->rq_export,
la.la_uid, la.la_gid));
}
/* copied from lov/lov_ea.c, just for debugging, will be removed later */
void mdt_dump_lmm(int level, const struct lov_mds_md *lmm, __u64 valid)
{
const struct lov_ost_data_v1 *lod;
__u32 lmm_magic = le32_to_cpu(lmm->lmm_magic);
__u16 count;
int i;
if (likely(!cfs_cdebug_show(level, DEBUG_SUBSYSTEM)))
return;
CDEBUG_LIMIT(level, "objid="DOSTID" magic=0x%08X pattern=%#X\n",
POSTID(&lmm->lmm_oi), lmm_magic,
le32_to_cpu(lmm->lmm_pattern));
/* No support for compound layouts yet */
if (lmm_magic != LOV_MAGIC_V1 && lmm_magic != LOV_MAGIC_V3)
return;
count = le16_to_cpu(((struct lov_user_md *)lmm)->lmm_stripe_count);
CDEBUG_LIMIT(level, "stripe_size=0x%x, %sstripe_count=0x%x\n",
le32_to_cpu(lmm->lmm_stripe_size),
count > LOV_MAX_STRIPE_COUNT ? "bad " : "", count);
/* If it's a directory or a released file, then there are
* no actual objects to print, so bail out.
*/
if (valid & OBD_MD_FLDIREA ||
le32_to_cpu(lmm->lmm_pattern) & LOV_PATTERN_F_RELEASED)
return;
if (unlikely(count > LOV_MAX_STRIPE_COUNT))
return;
for (i = 0, lod = lmm->lmm_objects; i < count; i++, lod++) {
struct ost_id oi;
ostid_le_to_cpu(&lod->l_ost_oi, &oi);
CDEBUG_LIMIT(level, "stripe %u idx %u subobj "DOSTID"\n",
i, le32_to_cpu(lod->l_ost_idx), POSTID(&oi));
}
}
void mdt_dump_lmv(unsigned int level, const union lmv_mds_md *lmv)
{
const struct lmv_mds_md_v1 *lmm1;
const struct lmv_foreign_md *lfm;
int i;
if (likely(!cfs_cdebug_show(level, DEBUG_SUBSYSTEM)))
return;
/* foreign LMV case */
lfm = &lmv->lmv_foreign_md;
if (le32_to_cpu(lfm->lfm_magic) == LMV_MAGIC_FOREIGN) {
CDEBUG_LIMIT(level,
"foreign magic 0x%08X, length %u, type %u, flags %u, value '%.*s'\n",
le32_to_cpu(lfm->lfm_magic),
le32_to_cpu(lfm->lfm_length),
le32_to_cpu(lfm->lfm_type),
le32_to_cpu(lfm->lfm_flags),
le32_to_cpu(lfm->lfm_length), lfm->lfm_value);
return;
}
lmm1 = &lmv->lmv_md_v1;
CDEBUG(level,
"magic 0x%08X, master %#X stripe_count %d hash_type %#x\n",
le32_to_cpu(lmm1->lmv_magic),
le32_to_cpu(lmm1->lmv_master_mdt_index),
le32_to_cpu(lmm1->lmv_stripe_count),
le32_to_cpu(lmm1->lmv_hash_type));
if (le32_to_cpu(lmm1->lmv_magic) == LMV_MAGIC_STRIPE)
return;
if (le32_to_cpu(lmm1->lmv_stripe_count) > LMV_MAX_STRIPE_COUNT)
return;
for (i = 0; i < le32_to_cpu(lmm1->lmv_stripe_count); i++) {
struct lu_fid fid;
fid_le_to_cpu(&fid, &lmm1->lmv_stripe_fids[i]);
CDEBUG(level, "idx %u subobj "DFID"\n", i, PFID(&fid));
}
}
/* Shrink and/or grow reply buffers */
int mdt_fix_reply(struct mdt_thread_info *info)
{
struct req_capsule *pill = info->mti_pill;
struct mdt_body *body;
int md_size, md_packed = 0;
int acl_size;
int rc = 0;
ENTRY;
body = req_capsule_server_get(pill, &RMF_MDT_BODY);
LASSERT(body != NULL);
if (body->mbo_valid & (OBD_MD_FLDIREA | OBD_MD_FLEASIZE |
OBD_MD_LINKNAME))
md_size = body->mbo_eadatasize;
else
md_size = 0;
acl_size = body->mbo_aclsize;
/* this replay - not send info to client */
if (info->mti_spec.no_create) {
md_size = 0;
acl_size = 0;
}
CDEBUG(D_INFO, "Shrink to md_size = %d cookie/acl_size = %d\n",
md_size, acl_size);
/*
&RMF_MDT_BODY,
&RMF_MDT_MD,
&RMF_ACL, or &RMF_LOGCOOKIES
(optional) &RMF_CAPA1,
(optional) &RMF_CAPA2,
(optional) something else
*/
/* MDT_MD buffer may be bigger than packed value, let's shrink all
* buffers before growing it
*/
if (info->mti_big_lov_used || info->mti_big_lmv_used) {
/* big_lmm buffer may be used even without packing the result
* into reply, just for internal server needs
*/
if (req_capsule_has_field(pill, &RMF_MDT_MD, RCL_SERVER))
md_packed = req_capsule_get_size(pill, &RMF_MDT_MD,
RCL_SERVER);
/* free big lmm if md_size is not needed */
if (md_size == 0 || md_packed == 0) {
info->mti_big_lov_used = 0;
info->mti_big_lmv_used = 0;
} else {
/* buffer must be allocated separately */
LASSERT(info->mti_attr.ma_lmm !=
req_capsule_server_get(pill, &RMF_MDT_MD));
req_capsule_shrink(pill, &RMF_MDT_MD, 0, RCL_SERVER);
md_packed = 0;
}
} else if (req_capsule_has_field(pill, &RMF_MDT_MD, RCL_SERVER)) {
req_capsule_shrink(pill, &RMF_MDT_MD, md_size, RCL_SERVER);
}
if (info->mti_big_acl_used) {
if (acl_size == 0)
info->mti_big_acl_used = 0;
else
req_capsule_shrink(pill, &RMF_ACL, 0, RCL_SERVER);
} else if (req_capsule_has_field(pill, &RMF_ACL, RCL_SERVER)) {
req_capsule_shrink(pill, &RMF_ACL, acl_size, RCL_SERVER);
} else if (req_capsule_has_field(pill, &RMF_LOGCOOKIES, RCL_SERVER)) {
req_capsule_shrink(pill, &RMF_LOGCOOKIES, acl_size, RCL_SERVER);
}
/* Shrink optional SECCTX buffer if it is not used */
if (req_capsule_has_field(pill, &RMF_FILE_SECCTX, RCL_SERVER) &&
req_capsule_get_size(pill, &RMF_FILE_SECCTX, RCL_SERVER) != 0 &&
!(body->mbo_valid & OBD_MD_SECCTX))
req_capsule_shrink(pill, &RMF_FILE_SECCTX, 0, RCL_SERVER);
/* Shrink optional ENCCTX buffer if it is not used */
if (req_capsule_has_field(pill, &RMF_FILE_ENCCTX, RCL_SERVER) &&
req_capsule_get_size(pill, &RMF_FILE_ENCCTX, RCL_SERVER) != 0 &&
!(body->mbo_valid & OBD_MD_ENCCTX))
req_capsule_shrink(pill, &RMF_FILE_ENCCTX, 0, RCL_SERVER);
/* Shrink optional default LMV buffer if it is not used */
if (req_capsule_has_field(pill, &RMF_DEFAULT_MDT_MD, RCL_SERVER) &&
req_capsule_get_size(pill, &RMF_DEFAULT_MDT_MD, RCL_SERVER) != 0 &&
!(body->mbo_valid & OBD_MD_DEFAULT_MEA))
req_capsule_shrink(pill, &RMF_DEFAULT_MDT_MD, 0, RCL_SERVER);
/*
* Some more field should be shrinked if needed.
* This should be done by those who added fields to reply message.
*/
/* Grow MD buffer if needed finally */
if (info->mti_big_lov_used || info->mti_big_lmv_used) {
void *lmm;
LASSERT(md_size > md_packed);
CDEBUG(D_INFO, "Enlarge reply buffer, need extra %d bytes\n",
md_size - md_packed);
rc = req_capsule_server_grow(pill, &RMF_MDT_MD, md_size);
if (rc) {
/* we can't answer with proper LOV EA, drop flags,
* the rc is also returned so this request is
* considered as failed
*/
body->mbo_valid &= ~(OBD_MD_FLDIREA | OBD_MD_FLEASIZE);
/* don't return transno along with error */
lustre_msg_set_transno(pill->rc_req->rq_repmsg, 0);
} else {
/* now we need to pack right LOV/LMV EA */
lmm = req_capsule_server_get(pill, &RMF_MDT_MD);
if (info->mti_attr.ma_valid & MA_LOV) {
LASSERT(req_capsule_get_size(pill, &RMF_MDT_MD,
RCL_SERVER) ==
info->mti_attr.ma_lmm_size);
memcpy(lmm, info->mti_attr.ma_lmm,
info->mti_attr.ma_lmm_size);
} else if (info->mti_attr.ma_valid & MA_LMV) {
LASSERT(req_capsule_get_size(pill, &RMF_MDT_MD,
RCL_SERVER) ==
info->mti_attr.ma_lmv_size);
memcpy(lmm, info->mti_attr.ma_lmv,
info->mti_attr.ma_lmv_size);
}
}
/* update mdt_max_mdsize so clients will be aware about that */
if (info->mti_mdt->mdt_max_mdsize < info->mti_attr.ma_lmm_size)
info->mti_mdt->mdt_max_mdsize =
info->mti_attr.ma_lmm_size;
info->mti_big_lov_used = 0;
info->mti_big_lmv_used = 0;
}
if (info->mti_big_acl_used) {
CDEBUG(D_INFO, "Enlarge reply ACL buffer to %d bytes\n",
acl_size);
rc = req_capsule_server_grow(pill, &RMF_ACL, acl_size);
if (rc) {
body->mbo_valid &= ~OBD_MD_FLACL;
} else {
void *acl = req_capsule_server_get(pill, &RMF_ACL);
memcpy(acl, info->mti_big_acl, acl_size);
}
info->mti_big_acl_used = 0;
}
RETURN(rc);
}
/* if object is dying, pack the lov/llog data,
* parameter info->mti_attr should be valid at this point!
* Also implements RAoLU policy
*/
int mdt_handle_last_unlink(struct mdt_thread_info *info, struct mdt_object *mo,
struct md_attr *ma)
{
struct mdt_body *repbody = NULL;
const struct lu_attr *la = &ma->ma_attr;
struct coordinator *cdt = &info->mti_mdt->mdt_coordinator;
int rc;
__u64 need = 0;
struct hsm_action_item hai = {
.hai_len = sizeof(hai),
.hai_action = HSMA_REMOVE,
.hai_extent.length = -1,
.hai_cookie = 0,
.hai_gid = 0,
};
int archive_id;
ENTRY;
if (mdt_info_req(info) != NULL) {
repbody = req_capsule_server_get(info->mti_pill, &RMF_MDT_BODY);
LASSERT(repbody != NULL);
} else {
CDEBUG(D_INFO, "not running in a request/reply context\n");
}
if ((ma->ma_valid & MA_INODE) && repbody != NULL)
mdt_pack_attr2body(info, repbody, la, mdt_object_fid(mo));
if (ma->ma_valid & MA_LOV) {
CERROR("No need in LOV EA upon unlink\n");
dump_stack();
}
if (repbody != NULL)
repbody->mbo_eadatasize = 0;
/* Only check unlinked and archived if RAoLU and upon last close */
if (!cdt->cdt_remove_archive_on_last_unlink ||
atomic_read(&mo->mot_open_count) != 0)
RETURN(0);
/* mdt_attr_get_complex will clear ma_valid, so check here first */
if ((ma->ma_valid & MA_INODE) && (ma->ma_attr.la_nlink != 0))
RETURN(0);
if ((ma->ma_valid & MA_HSM) && (!(ma->ma_hsm.mh_flags & HS_EXISTS)))
RETURN(0);
need |= (MA_INODE | MA_HSM) & ~ma->ma_valid;
if (need != 0) {
/* ma->ma_valid is missing either MA_INODE, MA_HSM, or both,
* try setting them
*/
ma->ma_need |= need;
rc = mdt_attr_get_complex(info, mo, ma);
if (rc) {
CERROR("%s: unable to fetch missing attributes of "DFID": rc = %d\n",
mdt_obd_name(info->mti_mdt),
PFID(mdt_object_fid(mo)), rc);
RETURN(0);
}
if (need & MA_INODE) {
if (ma->ma_valid & MA_INODE) {
if (ma->ma_attr.la_nlink != 0)
RETURN(0);
} else {
RETURN(0);
}
}
if (need & MA_HSM) {
if (ma->ma_valid & MA_HSM) {
if (!(ma->ma_hsm.mh_flags & HS_EXISTS))
RETURN(0);
} else {
RETURN(0);
}
}
}
/* RAoLU policy is active, last close on file has occured,
* file is unlinked, file is archived, so create remove request
* for copytool!
* If CDT is not running, requests will be logged for later.
*/
if (ma->ma_hsm.mh_arch_id != 0)
archive_id = ma->ma_hsm.mh_arch_id;
else
archive_id = cdt->cdt_default_archive_id;
hai.hai_fid = *mdt_object_fid(mo);
rc = mdt_agent_record_add(info->mti_env, info->mti_mdt, archive_id, 0,
&hai);
if (rc)
CERROR("%s: unable to add HSM remove request for "DFID": rc = %d\n",
mdt_obd_name(info->mti_mdt), PFID(mdt_object_fid(mo)),
rc);
RETURN(0);
}
static __u64 mdt_attr_valid_xlate(enum mds_attr_flags in,
struct mdt_reint_record *rr,
struct md_attr *ma)
{
__u64 out;
out = 0;
if (in & MDS_ATTR_MODE)
out |= LA_MODE;
if (in & MDS_ATTR_UID)
out |= LA_UID;
if (in & MDS_ATTR_GID)
out |= LA_GID;
if (in & MDS_ATTR_SIZE)
out |= LA_SIZE;
if (in & MDS_ATTR_BLOCKS)
out |= LA_BLOCKS;
if (in & MDS_ATTR_ATIME_SET)
out |= LA_ATIME;
if (in & MDS_ATTR_CTIME_SET)
out |= LA_CTIME;
if (in & MDS_ATTR_MTIME_SET)
out |= LA_MTIME;
if (in & MDS_ATTR_ATTR_FLAG)
out |= LA_FLAGS;
if (in & MDS_ATTR_KILL_SUID)
out |= LA_KILL_SUID;
if (in & MDS_ATTR_KILL_SGID)
out |= LA_KILL_SGID;
if (in & MDS_ATTR_PROJID)
out |= LA_PROJID;
if (in & MDS_ATTR_LSIZE)
out |= LA_LSIZE;
if (in & MDS_ATTR_LBLOCKS)
out |= LA_LBLOCKS;
if (in & MDS_ATTR_FROM_OPEN)
rr->rr_flags |= MRF_OPEN_TRUNC;
if (in & MDS_ATTR_OVERRIDE)
ma->ma_attr_flags |= MDS_OWNEROVERRIDE;
if (in & MDS_ATTR_FORCE)
ma->ma_attr_flags |= MDS_PERM_BYPASS;
in &= ~(MDS_ATTR_MODE | MDS_ATTR_UID | MDS_ATTR_GID | MDS_ATTR_PROJID |
MDS_ATTR_ATIME | MDS_ATTR_MTIME | MDS_ATTR_CTIME |
MDS_ATTR_ATIME_SET | MDS_ATTR_CTIME_SET | MDS_ATTR_MTIME_SET |
MDS_ATTR_SIZE | MDS_ATTR_BLOCKS | MDS_ATTR_ATTR_FLAG |
MDS_ATTR_FORCE | MDS_ATTR_KILL_SUID | MDS_ATTR_KILL_SGID |
MDS_ATTR_FROM_OPEN | MDS_ATTR_LSIZE | MDS_ATTR_LBLOCKS |
MDS_ATTR_OVERRIDE);
if (in != 0)
CDEBUG(D_INFO, "Unknown attr bits: %#llx\n", (u64)in);
return out;
}
/* unpacking */
int mdt_name_unpack(struct req_capsule *pill,
const struct req_msg_field *field,
struct lu_name *ln,
enum mdt_name_flags flags)
{
ln->ln_name = req_capsule_client_get(pill, field);
ln->ln_namelen = req_capsule_get_size(pill, field, RCL_CLIENT) - 1;
if (!lu_name_is_valid(ln)) {
ln->ln_name = NULL;
ln->ln_namelen = 0;
return -EPROTO;
}
if ((flags & MNF_FIX_ANON) &&
ln->ln_namelen == 1 && ln->ln_name[0] == '/') {
/* Newer (3.x) kernels use a name of "/" for the
* "anonymous" disconnected dentries from NFS
* filehandle conversion. See d_obtain_alias().
*/
ln->ln_name = NULL;
ln->ln_namelen = 0;
}
return 0;
}
static int mdt_file_secctx_unpack(struct req_capsule *pill,
const char **secctx_name,
void **secctx, size_t *secctx_size)
{
const char *name;
size_t name_size;
*secctx_name = NULL;
*secctx = NULL;
*secctx_size = 0;
if (!req_capsule_has_field(pill, &RMF_FILE_SECCTX_NAME, RCL_CLIENT) ||
!req_capsule_field_present(pill, &RMF_FILE_SECCTX_NAME, RCL_CLIENT))
return 0;
name_size = req_capsule_get_size(pill, &RMF_FILE_SECCTX_NAME,
RCL_CLIENT);
if (name_size == 0)
return 0;
if (name_size > XATTR_NAME_MAX + 1)
return -EPROTO;
name = req_capsule_client_get(pill, &RMF_FILE_SECCTX_NAME);
if (strnlen(name, name_size) != name_size - 1)
return -EPROTO;
if (!req_capsule_has_field(pill, &RMF_FILE_SECCTX, RCL_CLIENT) ||
!req_capsule_field_present(pill, &RMF_FILE_SECCTX, RCL_CLIENT))
return -EPROTO;
*secctx_name = name;
*secctx = req_capsule_client_get(pill, &RMF_FILE_SECCTX);
*secctx_size = req_capsule_get_size(pill, &RMF_FILE_SECCTX, RCL_CLIENT);
return 0;
}
static int mdt_file_encctx_unpack(struct req_capsule *pill,
void **encctx, size_t *encctx_size)
{
*encctx = NULL;
*encctx_size = 0;
if (!exp_connect_encrypt(pill->rc_req->rq_export))
return 0;
if (!req_capsule_has_field(pill, &RMF_FILE_ENCCTX, RCL_CLIENT) ||
!req_capsule_field_present(pill, &RMF_FILE_ENCCTX, RCL_CLIENT))
return -EPROTO;
*encctx_size = req_capsule_get_size(pill, &RMF_FILE_ENCCTX, RCL_CLIENT);
if (*encctx_size == 0)
return 0;
*encctx = req_capsule_client_get(pill, &RMF_FILE_ENCCTX);
return 0;
}
static int mdt_setattr_unpack_rec(struct mdt_thread_info *info)
{
struct lu_ucred *uc = mdt_ucred(info);
struct md_attr *ma = &info->mti_attr;
struct lu_attr *la = &ma->ma_attr;
struct req_capsule *pill = info->mti_pill;
struct mdt_reint_record *rr = &info->mti_rr;
struct mdt_rec_setattr *rec;
struct lu_nodemap *nodemap;
ENTRY;
BUILD_BUG_ON(sizeof(*rec) != sizeof(struct mdt_rec_reint));
rec = req_capsule_client_get(pill, &RMF_REC_REINT);
if (rec == NULL)
RETURN(-EFAULT);
/* This prior initialization is needed for old_init_ucred_reint() */
uc->uc_fsuid = rec->sa_fsuid;
uc->uc_fsgid = rec->sa_fsgid;
uc->uc_cap = CAP_EMPTY_SET;
ll_set_capability_u32(&uc->uc_cap, rec->sa_cap);
rr->rr_fid1 = &rec->sa_fid;
la->la_valid = mdt_attr_valid_xlate(rec->sa_valid, rr, ma);
la->la_mode = rec->sa_mode;
if (rec->sa_attr_flags & ~LUSTRE_FL_USER_VISIBLE) {
CDEBUG(D_INODE, "Unsupported flags: %x\n", rec->sa_attr_flags);
RETURN(-EOPNOTSUPP);
}
la->la_flags = rec->sa_attr_flags & LUSTRE_FL_USER_MODIFIABLE;
nodemap = nodemap_get_from_exp(info->mti_exp);
if (IS_ERR(nodemap))
RETURN(PTR_ERR(nodemap));
la->la_uid = nodemap_map_id(nodemap, NODEMAP_UID,
NODEMAP_CLIENT_TO_FS, rec->sa_uid);
la->la_gid = nodemap_map_id(nodemap, NODEMAP_GID,
NODEMAP_CLIENT_TO_FS, rec->sa_gid);
la->la_projid = nodemap_map_id(nodemap, NODEMAP_PROJID,
NODEMAP_CLIENT_TO_FS, rec->sa_projid);
uc->uc_suppgids[0] = nodemap_map_suppgid(nodemap, rec->sa_suppgid);
uc->uc_suppgids[1] = -1;
nodemap_putref(nodemap);
la->la_size = rec->sa_size;
la->la_blocks = rec->sa_blocks;
la->la_ctime = rec->sa_ctime;
la->la_atime = rec->sa_atime;
la->la_mtime = rec->sa_mtime;
ma->ma_valid = MA_INODE;
ma->ma_attr_flags |= rec->sa_bias & (MDS_CLOSE_INTENT |
MDS_DATA_MODIFIED | MDS_TRUNC_KEEP_LEASE |
MDS_PCC_ATTACH | MDS_CLOSE_LAYOUT_SWAP_HSM);
RETURN(0);
}
static int mdt_close_handle_unpack(struct mdt_thread_info *info)
{
struct req_capsule *pill = info->mti_pill;
struct mdt_ioepoch *ioepoch;
ENTRY;
if (req_capsule_get_size(pill, &RMF_MDT_EPOCH, RCL_CLIENT))
ioepoch = req_capsule_client_get(pill, &RMF_MDT_EPOCH);
else
ioepoch = NULL;
if (ioepoch == NULL)
RETURN(-EPROTO);
info->mti_open_handle = ioepoch->mio_open_handle;
RETURN(0);
}
static inline int mdt_dlmreq_unpack(struct mdt_thread_info *info)
{
struct req_capsule *pill = info->mti_pill;
if (!info->mti_intent_lock &&
req_capsule_get_size(pill, &RMF_DLM_REQ, RCL_CLIENT)) {
info->mti_dlm_req = req_capsule_client_get(pill, &RMF_DLM_REQ);
if (info->mti_dlm_req == NULL)
RETURN(-EFAULT);
}
RETURN(0);
}
static int mdt_setattr_unpack(struct mdt_thread_info *info)
{
struct mdt_reint_record *rr = &info->mti_rr;
struct md_attr *ma = &info->mti_attr;
struct req_capsule *pill = info->mti_pill;
int rc;
ENTRY;
rc = mdt_setattr_unpack_rec(info);
if (rc)
RETURN(rc);
if (req_capsule_field_present(pill, &RMF_EADATA, RCL_CLIENT)) {
rr->rr_eadata = req_capsule_client_get(pill, &RMF_EADATA);
rr->rr_eadatalen = req_capsule_get_size(pill, &RMF_EADATA,
RCL_CLIENT);
if (rr->rr_eadatalen > 0) {
const struct lmv_user_md *lum;
lum = rr->rr_eadata;
/* Sigh ma_valid(from req) does not indicate whether
* it will set LOV/LMV EA, so we have to check magic
*/
if (le32_to_cpu(lum->lum_magic) == LMV_USER_MAGIC) {
ma->ma_valid |= MA_LMV;
ma->ma_lmv = (void *)rr->rr_eadata;
ma->ma_lmv_size = rr->rr_eadatalen;
} else {
ma->ma_valid |= MA_LOV;
ma->ma_lmm = (void *)rr->rr_eadata;
ma->ma_lmm_size = rr->rr_eadatalen;
}
}
}
rc = mdt_dlmreq_unpack(info);
RETURN(rc);
}
static int mdt_close_intent_unpack(struct mdt_thread_info *info)
{
struct md_attr *ma = &info->mti_attr;
struct req_capsule *pill = info->mti_pill;
ENTRY;
if (!(ma->ma_attr_flags & MDS_CLOSE_INTENT))
RETURN(0);
req_capsule_extend(pill, &RQF_MDS_CLOSE_INTENT);
if (!(req_capsule_has_field(pill, &RMF_CLOSE_DATA, RCL_CLIENT) &&
req_capsule_field_present(pill, &RMF_CLOSE_DATA, RCL_CLIENT)))
RETURN(-EFAULT);
RETURN(0);
}
int mdt_close_unpack(struct mdt_thread_info *info)
{
int rc;
ENTRY;
rc = mdt_close_handle_unpack(info);
if (rc)
RETURN(rc);
rc = mdt_setattr_unpack_rec(info);
if (rc)
RETURN(rc);
rc = mdt_close_intent_unpack(info);
if (rc)
RETURN(rc);
RETURN(mdt_init_ucred_reint(info));
}
static int mdt_create_unpack(struct mdt_thread_info *info)
{
struct lu_ucred *uc = mdt_ucred(info);
struct mdt_rec_create *rec;
struct lu_attr *attr = &info->mti_attr.ma_attr;
struct mdt_reint_record *rr = &info->mti_rr;
struct req_capsule *pill = info->mti_pill;
struct md_op_spec *sp = &info->mti_spec;
struct lu_nodemap *nodemap;
int rc;
ENTRY;
BUILD_BUG_ON(sizeof(*rec) != sizeof(struct mdt_rec_reint));
rec = req_capsule_client_get(pill, &RMF_REC_REINT);
if (rec == NULL)
RETURN(-EFAULT);
/* This prior initialization is needed for old_init_ucred_reint() */
uc->uc_fsuid = rec->cr_fsuid;
uc->uc_fsgid = rec->cr_fsgid;
uc->uc_cap = CAP_EMPTY_SET;
ll_set_capability_u32(&uc->uc_cap, rec->cr_cap);
nodemap = nodemap_get_from_exp(info->mti_exp);
if (IS_ERR(nodemap))
RETURN(PTR_ERR(nodemap));
uc->uc_suppgids[0] = nodemap_map_suppgid(nodemap, rec->cr_suppgid1);
uc->uc_suppgids[1] = -1;
nodemap_putref(nodemap);
uc->uc_umask = rec->cr_umask;
rr->rr_fid1 = &rec->cr_fid1;
rr->rr_fid2 = &rec->cr_fid2;
rr->rr_layout_ver = rec->cr_layout_ver;
attr->la_mode = rec->cr_mode;
attr->la_rdev = rec->cr_rdev;
attr->la_uid = rec->cr_fsuid;
attr->la_gid = rec->cr_fsgid;
attr->la_ctime = rec->cr_time;
attr->la_mtime = rec->cr_time;
attr->la_atime = rec->cr_time;
attr->la_valid = LA_MODE | LA_RDEV | LA_UID | LA_GID | LA_TYPE |
LA_CTIME | LA_MTIME | LA_ATIME;
memset(&sp->u, 0, sizeof(sp->u));
sp->sp_cr_flags = get_mrc_cr_flags(rec);
rc = mdt_name_unpack(pill, &RMF_NAME, &rr->rr_name, 0);
if (rc < 0)
RETURN(rc);
if (S_ISLNK(attr->la_mode)) {
const char *tgt = NULL;
int sz;
req_capsule_extend(pill, &RQF_MDS_REINT_CREATE_SYM);
sz = req_capsule_get_size(pill, &RMF_SYMTGT, RCL_CLIENT);
if (sz) {
tgt = req_capsule_client_get(pill, &RMF_SYMTGT);
sp->u.sp_symname.ln_name = tgt;
sp->u.sp_symname.ln_namelen = sz - 1; /* skip NUL */
}
if (tgt == NULL)
RETURN(-EFAULT);
} else {
if (!info->mti_intent_lock)
req_capsule_extend(pill, &RQF_MDS_REINT_CREATE_ACL);
rr->rr_eadatalen = req_capsule_get_size(pill, &RMF_EADATA,
RCL_CLIENT);
if (rr->rr_eadatalen > 0) {
sp->no_create = !!req_is_replay(mdt_info_req(info));
if (S_ISDIR(attr->la_mode)) {
sp->u.sp_ea.eadata =
req_capsule_client_get(pill,
&RMF_EADATA);
sp->u.sp_ea.eadatalen = rr->rr_eadatalen;
sp->sp_cr_flags |= MDS_OPEN_HAS_EA;
}
}
if (S_ISDIR(attr->la_mode)) {
struct obd_export *exp = mdt_info_req(info)->rq_export;
sp->sp_dmv_imp_inherit =
info->mti_mdt->mdt_enable_dmv_implicit_inherit;
if (OCD_HAS_FLAG2(&exp->exp_connect_data,
DMV_IMP_INHERIT)) {
if ((sp->sp_cr_flags & MDS_OPEN_DEFAULT_LMV) &&
!(sp->sp_cr_flags & MDS_OPEN_HAS_EA))
RETURN(-EPROTO);
} else if (sp->sp_cr_flags & MDS_OPEN_DEFAULT_LMV) {
RETURN(-EPROTO);
}
}
}
rc = mdt_file_secctx_unpack(pill, &sp->sp_cr_file_secctx_name,
&sp->sp_cr_file_secctx,
&sp->sp_cr_file_secctx_size);
if (rc < 0)
RETURN(rc);
rc = mdt_file_encctx_unpack(pill, &sp->sp_cr_file_encctx,
&sp->sp_cr_file_encctx_size);
if (rc < 0)
RETURN(rc);
rc = req_check_sepol(pill);
if (rc)
RETURN(rc);
rc = mdt_dlmreq_unpack(info);
RETURN(rc);
}
static int mdt_link_unpack(struct mdt_thread_info *info)
{
struct lu_ucred *uc = mdt_ucred(info);
struct mdt_rec_link *rec;
struct lu_attr *attr = &info->mti_attr.ma_attr;
struct mdt_reint_record *rr = &info->mti_rr;
struct req_capsule *pill = info->mti_pill;
struct lu_nodemap *nodemap;
int rc;
ENTRY;
BUILD_BUG_ON(sizeof(*rec) != sizeof(struct mdt_rec_reint));
rec = req_capsule_client_get(pill, &RMF_REC_REINT);
if (rec == NULL)
RETURN(-EFAULT);
/* This prior initialization is needed for old_init_ucred_reint() */
uc->uc_fsuid = rec->lk_fsuid;
uc->uc_fsgid = rec->lk_fsgid;
uc->uc_cap = CAP_EMPTY_SET;
ll_set_capability_u32(&uc->uc_cap, rec->lk_cap);
nodemap = nodemap_get_from_exp(info->mti_exp);
if (IS_ERR(nodemap))
RETURN(PTR_ERR(nodemap));
uc->uc_suppgids[0] = nodemap_map_suppgid(nodemap, rec->lk_suppgid1);
uc->uc_suppgids[1] = nodemap_map_suppgid(nodemap, rec->lk_suppgid2);
nodemap_putref(nodemap);
attr->la_uid = rec->lk_fsuid;
attr->la_gid = rec->lk_fsgid;
rr->rr_fid1 = &rec->lk_fid1;
rr->rr_fid2 = &rec->lk_fid2;
attr->la_ctime = rec->lk_time;
attr->la_mtime = rec->lk_time;
attr->la_valid = LA_UID | LA_GID | LA_CTIME | LA_MTIME;
rc = mdt_name_unpack(pill, &RMF_NAME, &rr->rr_name, 0);
if (rc < 0)
RETURN(rc);
rc = req_check_sepol(pill);
if (rc)
RETURN(rc);
rc = mdt_dlmreq_unpack(info);
RETURN(rc);
}
static int mdt_unlink_unpack(struct mdt_thread_info *info)
{
struct lu_ucred *uc = mdt_ucred(info);
struct mdt_rec_unlink *rec;
struct lu_attr *attr = &info->mti_attr.ma_attr;
struct mdt_reint_record *rr = &info->mti_rr;
struct req_capsule *pill = info->mti_pill;
struct lu_nodemap *nodemap;
int rc;
ENTRY;
BUILD_BUG_ON(sizeof(*rec) != sizeof(struct mdt_rec_reint));
rec = req_capsule_client_get(pill, &RMF_REC_REINT);
if (rec == NULL)
RETURN(-EFAULT);
/* This prior initialization is needed for old_init_ucred_reint() */
uc->uc_fsuid = rec->ul_fsuid;
uc->uc_fsgid = rec->ul_fsgid;
uc->uc_cap = CAP_EMPTY_SET;
ll_set_capability_u32(&uc->uc_cap, rec->ul_cap);
nodemap = nodemap_get_from_exp(info->mti_exp);
if (IS_ERR(nodemap))
RETURN(PTR_ERR(nodemap));
uc->uc_suppgids[0] = nodemap_map_suppgid(nodemap, rec->ul_suppgid1);
uc->uc_suppgids[1] = -1;
nodemap_putref(nodemap);
attr->la_uid = rec->ul_fsuid;
attr->la_gid = rec->ul_fsgid;
rr->rr_fid1 = &rec->ul_fid1;
rr->rr_fid2 = &rec->ul_fid2;
attr->la_ctime = rec->ul_time;
attr->la_mtime = rec->ul_time;
attr->la_mode = rec->ul_mode;
attr->la_valid = LA_UID | LA_GID | LA_CTIME | LA_MTIME | LA_MODE;
if (rec->ul_bias & MDS_FID_OP)
info->mti_spec.sp_cr_flags |= MDS_OP_WITH_FID;
else
info->mti_spec.sp_cr_flags &= ~MDS_OP_WITH_FID;
rc = mdt_name_unpack(pill, &RMF_NAME, &rr->rr_name, 0);
if (rc < 0)
RETURN(rc);
info->mti_spec.no_create = !!req_is_replay(mdt_info_req(info));
rc = req_check_sepol(pill);
if (rc)
RETURN(rc);
rc = mdt_dlmreq_unpack(info);
RETURN(rc);
}
static int mdt_rmentry_unpack(struct mdt_thread_info *info)
{
info->mti_spec.sp_rm_entry = 1;
return mdt_unlink_unpack(info);
}
static int mdt_rename_unpack(struct mdt_thread_info *info)
{
struct lu_ucred *uc = mdt_ucred(info);
struct mdt_rec_rename *rec;
struct lu_attr *attr = &info->mti_attr.ma_attr;
struct mdt_reint_record *rr = &info->mti_rr;
struct req_capsule *pill = info->mti_pill;
struct md_op_spec *spec = &info->mti_spec;
struct lu_nodemap *nodemap;
int rc;
ENTRY;
BUILD_BUG_ON(sizeof(*rec) != sizeof(struct mdt_rec_reint));
rec = req_capsule_client_get(pill, &RMF_REC_REINT);
if (rec == NULL)
RETURN(-EFAULT);
/* This prior initialization is needed for old_init_ucred_reint() */
uc->uc_fsuid = rec->rn_fsuid;
uc->uc_fsgid = rec->rn_fsgid;
uc->uc_cap = CAP_EMPTY_SET;
ll_set_capability_u32(&uc->uc_cap, rec->rn_cap);
nodemap = nodemap_get_from_exp(info->mti_exp);
if (IS_ERR(nodemap))
RETURN(PTR_ERR(nodemap));
uc->uc_suppgids[0] = nodemap_map_suppgid(nodemap, rec->rn_suppgid1);
uc->uc_suppgids[1] = nodemap_map_suppgid(nodemap, rec->rn_suppgid2);
nodemap_putref(nodemap);
attr->la_uid = rec->rn_fsuid;
attr->la_gid = rec->rn_fsgid;
rr->rr_fid1 = &rec->rn_fid1;
rr->rr_fid2 = &rec->rn_fid2;
attr->la_ctime = rec->rn_time;
attr->la_mtime = rec->rn_time;
/* rename_tgt contains the mode already */
attr->la_mode = rec->rn_mode;
attr->la_valid = LA_UID | LA_GID | LA_CTIME | LA_MTIME | LA_MODE;
rc = mdt_name_unpack(pill, &RMF_NAME, &rr->rr_name, 0);
if (rc < 0)
RETURN(rc);
rc = mdt_name_unpack(pill, &RMF_SYMTGT, &rr->rr_tgt_name, 0);
if (rc < 0)
RETURN(rc);
spec->no_create = !!req_is_replay(mdt_info_req(info));
rc = req_check_sepol(pill);
if (rc)
RETURN(rc);
rc = mdt_dlmreq_unpack(info);
RETURN(rc);
}
static int mdt_migrate_unpack(struct mdt_thread_info *info)
{
struct lu_ucred *uc = mdt_ucred(info);
struct mdt_rec_rename *rec;
struct lu_attr *attr = &info->mti_attr.ma_attr;
struct mdt_reint_record *rr = &info->mti_rr;
struct req_capsule *pill = info->mti_pill;
struct md_op_spec *spec = &info->mti_spec;
struct lu_nodemap *nodemap;
int rc;
ENTRY;
BUILD_BUG_ON(sizeof(*rec) != sizeof(struct mdt_rec_reint));
rec = req_capsule_client_get(pill, &RMF_REC_REINT);
if (rec == NULL)
RETURN(-EFAULT);
/* This prior initialization is needed for old_init_ucred_reint() */
uc->uc_fsuid = rec->rn_fsuid;
uc->uc_fsgid = rec->rn_fsgid;
uc->uc_cap = CAP_EMPTY_SET;
ll_set_capability_u32(&uc->uc_cap, rec->rn_cap);
nodemap = nodemap_get_from_exp(info->mti_exp);
if (IS_ERR(nodemap))
RETURN(PTR_ERR(nodemap));
uc->uc_suppgids[0] = nodemap_map_suppgid(nodemap, rec->rn_suppgid1);
uc->uc_suppgids[1] = nodemap_map_suppgid(nodemap, rec->rn_suppgid2);
nodemap_putref(nodemap);
attr->la_uid = rec->rn_fsuid;
attr->la_gid = rec->rn_fsgid;
rr->rr_fid1 = &rec->rn_fid1;
rr->rr_fid2 = &rec->rn_fid2;
attr->la_ctime = rec->rn_time;
attr->la_mtime = rec->rn_time;
/* rename_tgt contains the mode already */
attr->la_mode = rec->rn_mode;
attr->la_valid = LA_UID | LA_GID | LA_CTIME | LA_MTIME | LA_MODE;
spec->sp_cr_flags = 0;
rc = mdt_name_unpack(pill, &RMF_NAME, &rr->rr_name, 0);
if (rc < 0)
RETURN(rc);
if (rec->rn_bias & MDS_CLOSE_MIGRATE) {
rc = mdt_close_handle_unpack(info);
if (rc)
RETURN(rc);
spec->sp_migrate_close = 1;
} else {
spec->sp_migrate_close = 0;
}
spec->sp_migrate_nsonly = !!(rec->rn_bias & MDS_MIGRATE_NSONLY);
/* lustre version > 2.11 migration packs lum */
if (req_capsule_has_field(pill, &RMF_EADATA, RCL_CLIENT)) {
if (!req_capsule_field_present(pill, &RMF_EADATA, RCL_CLIENT))
/* old client doesn't provide lum. */
RETURN(-EOPNOTSUPP);
rr->rr_eadatalen = req_capsule_get_size(pill, &RMF_EADATA,
RCL_CLIENT);
if (rr->rr_eadatalen > 0) {
struct lmv_user_md_v1 *lum;
s32 stripe_count;
lum = req_capsule_client_get(pill, &RMF_EADATA);
stripe_count = le32_to_cpu(lum->lum_stripe_count);
if (rr->rr_eadatalen > XATTR_SIZE_MAX ||
stripe_count > LMV_MAX_STRIPE_COUNT ||
stripe_count < LMV_OVERSTRIPE_COUNT_MAX)
RETURN(-EOVERFLOW);
spec->u.sp_ea.eadatalen = rr->rr_eadatalen;
spec->u.sp_ea.eadata = rr->rr_eadata = lum;
spec->sp_cr_flags |= MDS_OPEN_HAS_EA;
}
}
spec->no_create = !!req_is_replay(mdt_info_req(info));
rc = mdt_dlmreq_unpack(info);
RETURN(rc);
}
/*
* please see comment above LOV_MAGIC_V1_DEFINED
*/
void mdt_fix_lov_magic(struct mdt_thread_info *info, void *eadata)
{
struct lov_user_md_v1 *v1 = eadata;
LASSERT(v1);
if (unlikely(req_is_replay(mdt_info_req(info)))) {
if ((v1->lmm_magic & LOV_MAGIC_MASK) == LOV_MAGIC_MAGIC)
v1->lmm_magic |= LOV_MAGIC_DEFINED;
else if ((v1->lmm_magic & __swab32(LOV_MAGIC_MAGIC)) ==
__swab32(LOV_MAGIC_MAGIC))
v1->lmm_magic |= __swab32(LOV_MAGIC_DEFINED);
}
}
static int mdt_open_unpack(struct mdt_thread_info *info)
{
struct lu_ucred *uc = mdt_ucred(info);
struct mdt_rec_create *rec;
struct lu_attr *attr = &info->mti_attr.ma_attr;
struct req_capsule *pill = info->mti_pill;
struct mdt_reint_record *rr = &info->mti_rr;
struct ptlrpc_request *req = mdt_info_req(info);
struct md_op_spec *sp = &info->mti_spec;
struct lu_nodemap *nodemap;
int rc;
ENTRY;
BUILD_BUG_ON(sizeof(struct mdt_rec_create) !=
sizeof(struct mdt_rec_reint));
rec = req_capsule_client_get(pill, &RMF_REC_REINT);
if (rec == NULL)
RETURN(-EFAULT);
/* This prior initialization is needed for old_init_ucred_reint() */
uc->uc_fsuid = rec->cr_fsuid;
uc->uc_fsgid = rec->cr_fsgid;
uc->uc_cap = CAP_EMPTY_SET;
ll_set_capability_u32(&uc->uc_cap, rec->cr_cap);
nodemap = nodemap_get_from_exp(info->mti_exp);
if (IS_ERR(nodemap))
RETURN(PTR_ERR(nodemap));
uc->uc_suppgids[0] = nodemap_map_suppgid(nodemap, rec->cr_suppgid1);
uc->uc_suppgids[1] = nodemap_map_suppgid(nodemap, rec->cr_suppgid2);
nodemap_putref(nodemap);
uc->uc_umask = rec->cr_umask;
rr->rr_fid1 = &rec->cr_fid1;
rr->rr_fid2 = &rec->cr_fid2;
rr->rr_layout_ver = rec->cr_layout_ver;
rr->rr_open_handle = &rec->cr_open_handle_old;
attr->la_mode = rec->cr_mode;
attr->la_rdev = rec->cr_rdev;
attr->la_uid = rec->cr_fsuid;
attr->la_gid = rec->cr_fsgid;
attr->la_ctime = rec->cr_time;
attr->la_mtime = rec->cr_time;
attr->la_atime = rec->cr_time;
attr->la_valid = LA_MODE | LA_RDEV | LA_UID | LA_GID |
LA_CTIME | LA_MTIME | LA_ATIME;
memset(&info->mti_spec.u, 0, sizeof(info->mti_spec.u));
info->mti_spec.sp_cr_flags = get_mrc_cr_flags(rec);
/* Do not trigger ASSERTION if client miss to set such flags. */
if (unlikely(info->mti_spec.sp_cr_flags == 0))
RETURN(-EPROTO);
info->mti_cross_ref = !!(rec->cr_bias & MDS_CROSS_REF);
mdt_name_unpack(pill, &RMF_NAME, &rr->rr_name, MNF_FIX_ANON);
if (req_capsule_field_present(pill, &RMF_EADATA, RCL_CLIENT)) {
rr->rr_eadatalen = req_capsule_get_size(pill, &RMF_EADATA,
RCL_CLIENT);
if (rr->rr_eadatalen > 0) {
rr->rr_eadata = req_capsule_client_get(pill,
&RMF_EADATA);
sp->u.sp_ea.eadatalen = rr->rr_eadatalen;
sp->u.sp_ea.eadata = rr->rr_eadata;
sp->sp_archive_id = rec->cr_archive_id;
sp->no_create = !!req_is_replay(req);
mdt_fix_lov_magic(info, rr->rr_eadata);
}
/*
* Client default md_size may be 0 right after client start,
* until all osc are connected, set here just some reasonable
* value to prevent misbehavior.
*/
if (rr->rr_eadatalen == 0 &&
!(info->mti_spec.sp_cr_flags & MDS_OPEN_DELAY_CREATE))
rr->rr_eadatalen = MIN_MD_SIZE;
}
rc = mdt_file_secctx_unpack(pill, &sp->sp_cr_file_secctx_name,
&sp->sp_cr_file_secctx,
&sp->sp_cr_file_secctx_size);
if (rc < 0)
RETURN(rc);
rc = mdt_file_encctx_unpack(pill, &sp->sp_cr_file_encctx,
&sp->sp_cr_file_encctx_size);
if (rc < 0)
RETURN(rc);
rc = req_check_sepol(pill);
if (rc)
RETURN(rc);
RETURN(rc);
}
static int mdt_setxattr_unpack(struct mdt_thread_info *info)
{
struct mdt_reint_record *rr = &info->mti_rr;
struct lu_ucred *uc = mdt_ucred(info);
struct lu_attr *attr = &info->mti_attr.ma_attr;
struct req_capsule *pill = info->mti_pill;
struct mdt_rec_setxattr *rec;
struct lu_nodemap *nodemap;
int rc;
ENTRY;
BUILD_BUG_ON(sizeof(struct mdt_rec_setxattr) !=
sizeof(struct mdt_rec_reint));
rec = req_capsule_client_get(pill, &RMF_REC_REINT);
if (rec == NULL)
RETURN(-EFAULT);
/* This prior initialization is needed for old_init_ucred_reint() */
uc->uc_fsuid = rec->sx_fsuid;
uc->uc_fsgid = rec->sx_fsgid;
uc->uc_cap = CAP_EMPTY_SET;
ll_set_capability_u32(&uc->uc_cap, rec->sx_cap);
nodemap = nodemap_get_from_exp(info->mti_exp);
if (IS_ERR(nodemap))
RETURN(PTR_ERR(nodemap));
uc->uc_suppgids[0] = nodemap_map_suppgid(nodemap, rec->sx_suppgid1);
uc->uc_suppgids[1] = -1;
nodemap_putref(nodemap);
rr->rr_opcode = rec->sx_opcode;
rr->rr_fid1 = &rec->sx_fid;
attr->la_valid = rec->sx_valid;
attr->la_ctime = rec->sx_time;
attr->la_size = rec->sx_size;
attr->la_flags = rec->sx_flags;
rc = mdt_name_unpack(pill, &RMF_NAME, &rr->rr_name, 0);
if (rc < 0)
RETURN(rc);
if (req_capsule_field_present(pill, &RMF_EADATA, RCL_CLIENT)) {
rr->rr_eadatalen = req_capsule_get_size(pill, &RMF_EADATA,
RCL_CLIENT);
if (rr->rr_eadatalen > info->mti_mdt->mdt_max_ea_size)
RETURN(-E2BIG);
if (rr->rr_eadatalen > 0) {
rr->rr_eadata = req_capsule_client_get(pill,
&RMF_EADATA);
if (rr->rr_eadata == NULL)
RETURN(-EFAULT);
} else {
rr->rr_eadata = NULL;
}
} else if (!(attr->la_valid & OBD_MD_FLXATTRRM)) {
CDEBUG(D_INFO, "no xattr data supplied\n");
RETURN(-EFAULT);
}
rc = req_check_sepol(pill);
if (rc)
RETURN(rc);
if (mdt_dlmreq_unpack(info) < 0)
RETURN(-EPROTO);
RETURN(0);
}
static int mdt_resync_unpack(struct mdt_thread_info *info)
{
struct obd_export *exp = mdt_info_req(info)->rq_export;
struct req_capsule *pill = info->mti_pill;
struct mdt_reint_record *rr = &info->mti_rr;
struct lu_ucred *uc = mdt_ucred(info);
struct mdt_rec_resync *rec;
ENTRY;
BUILD_BUG_ON(sizeof(*rec) != sizeof(struct mdt_rec_reint));
rec = req_capsule_client_get(pill, &RMF_REC_REINT);
if (rec == NULL)
RETURN(-EFAULT);
/* This prior initialization is needed for old_init_ucred_reint() */
uc->uc_fsuid = rec->rs_fsuid;
uc->uc_fsgid = rec->rs_fsgid;
uc->uc_cap = CAP_EMPTY_SET;
ll_set_capability_u32(&uc->uc_cap, rec->rs_cap);
rr->rr_fid1 = &rec->rs_fid;
if (exp_connect_mirror_id_fix(exp))
rr->rr_mirror_id = (__u16)rec->rs_mirror_id_new;
else
rr->rr_mirror_id = rec->rs_mirror_id_old;
/* cookie doesn't need to be swapped but it has been swapped
* in lustre_swab_mdt_rec_reint() as rr_mtime, so here it needs
* restoring.
*/
if (req_capsule_req_need_swab(pill))
__swab64s(&rec->rs_lease_handle.cookie);
rr->rr_lease_handle = &rec->rs_lease_handle;
RETURN(mdt_dlmreq_unpack(info));
}
typedef int (*reint_unpacker)(struct mdt_thread_info *info);
static reint_unpacker mdt_reint_unpackers[REINT_MAX] = {
[REINT_SETATTR] = mdt_setattr_unpack,
[REINT_CREATE] = mdt_create_unpack,
[REINT_LINK] = mdt_link_unpack,
[REINT_UNLINK] = mdt_unlink_unpack,
[REINT_RENAME] = mdt_rename_unpack,
[REINT_OPEN] = mdt_open_unpack,
[REINT_SETXATTR] = mdt_setxattr_unpack,
[REINT_RMENTRY] = mdt_rmentry_unpack,
[REINT_MIGRATE] = mdt_migrate_unpack,
[REINT_RESYNC] = mdt_resync_unpack,
};
int mdt_reint_unpack(struct mdt_thread_info *info, __u32 op)
{
int rc;
ENTRY;
memset(&info->mti_rr, 0, sizeof(info->mti_rr));
if (op < REINT_MAX && mdt_reint_unpackers[op] != NULL) {
info->mti_rr.rr_opcode = op;
rc = mdt_reint_unpackers[op](info);
} else {
rc = -EFAULT;
CERROR("%s: Unexpected opcode %d: rc = %d\n",
mdt_obd_name(info->mti_mdt), op, rc);
}
RETURN(rc);
}
int mdt_pack_secctx_in_reply(struct mdt_thread_info *info,
struct mdt_object *child)
{
char *secctx_name;
struct lu_buf *buffer;
struct mdt_body *repbody;
struct req_capsule *pill = info->mti_pill;
int rc = 0;
if (req_capsule_has_field(pill, &RMF_FILE_SECCTX, RCL_SERVER) &&
req_capsule_get_size(pill, &RMF_FILE_SECCTX, RCL_SERVER) != 0) {
secctx_name =
req_capsule_client_get(pill, &RMF_FILE_SECCTX_NAME);
buffer = &info->mti_buf;
/* fill reply buffer with security context now */
buffer->lb_len = req_capsule_get_size(pill, &RMF_FILE_SECCTX,
RCL_SERVER);
buffer->lb_buf = req_capsule_server_get(info->mti_pill,
&RMF_FILE_SECCTX);
rc = mo_xattr_get(info->mti_env, mdt_object_child(child),
buffer, secctx_name);
if (rc >= 0) {
CDEBUG(D_SEC,
"found security context of size %d for "DFID"\n",
rc, PFID(mdt_object_fid(child)));
repbody = req_capsule_server_get(pill, &RMF_MDT_BODY);
repbody->mbo_valid |= OBD_MD_SECCTX;
if (rc < buffer->lb_len)
req_capsule_shrink(pill, &RMF_FILE_SECCTX, rc,
RCL_SERVER);
rc = 0;
} else {
CDEBUG(D_SEC,
"security context not found for "DFID": rc = %d\n",
PFID(mdt_object_fid(child)), rc);
req_capsule_shrink(pill, &RMF_FILE_SECCTX, 0,
RCL_SERVER);
/* handling -ENOENT is important because it may change
* object state in DNE env dropping LOHA_EXISTS flag,
* it is important to return that to the caller.
* Check LU-13115 for details.
*/
if (rc != -ENOENT)
rc = 0;
}
}
return rc;
}
/* check whether two FIDs belong to different MDTs.
* \retval 1 on different MDTs.
* 0 on the same MDT.
* -ve on error
*/
int mdt_fids_different_target(struct mdt_thread_info *info,
const struct lu_fid *fid1,
const struct lu_fid *fid2)
{
const struct lu_env *env = info->mti_env;
struct mdt_device *mdt = info->mti_mdt;
struct lu_seq_range *range = &info->mti_range;
struct seq_server_site *ss;
__u32 index1, index2;
int rc;
if (fid_seq(fid1) == fid_seq(fid2))
return 0;
ss = mdt->mdt_lu_dev.ld_site->ld_seq_site;
range->lsr_flags = LU_SEQ_RANGE_MDT;
rc = fld_server_lookup(env, ss->ss_server_fld, fid1->f_seq, range);
if (rc)
return rc;
index1 = range->lsr_index;
rc = fld_server_lookup(env, ss->ss_server_fld, fid2->f_seq, range);
if (rc)
return rc;
index2 = range->lsr_index;
return index1 != index2;
}
/**
* mdt_is_remote_object() - Check whether @child is remote object on @parent.
* @info: thread environment
* @parent: parent object, it's the same as child object in getattr_by_fid
* @child: child object
*
* Return:
* * %1 is remote object.
* * %0 isn't remote object.
* * %negative error code
*/
int mdt_is_remote_object(struct mdt_thread_info *info,
struct mdt_object *parent,
struct mdt_object *child)
{
struct lu_buf *buf = &info->mti_big_buf;
struct linkea_data ldata = { NULL };
struct link_ea_header *leh;
struct link_ea_entry *lee;
struct lu_name name;
struct lu_fid pfid;
int reclen;
int i;
int rc;
ENTRY;
if (fid_is_root(mdt_object_fid(child)))
RETURN(0);
if (likely(parent != child)) {
if (mdt_object_remote(parent) ^ mdt_object_remote(child))
RETURN(1);
if (!mdt_object_remote(parent) && !mdt_object_remote(child))
RETURN(0);
rc = mdt_fids_different_target(info, mdt_object_fid(parent),
mdt_object_fid(child));
RETURN(rc);
}
/* client < 2.13.52 getattr_by_fid parent and child are the same */
buf = lu_buf_check_and_alloc(buf, PATH_MAX);
if (!buf->lb_buf)
RETURN(-ENOMEM);
ldata.ld_buf = buf;
rc = mdt_links_read(info, child, &ldata);
/* can't read linkea, just assume it's remote object */
if (rc == -ENOENT || rc == -ENODATA)
RETURN(1);
if (rc)
RETURN(rc);
leh = buf->lb_buf;
lee = (struct link_ea_entry *)(leh + 1);
for (i = 0; i < leh->leh_reccount; i++) {
linkea_entry_unpack(lee, &reclen, &name, &pfid);
lee = (struct link_ea_entry *) ((char *)lee + reclen);
rc = mdt_fids_different_target(info, &pfid,
mdt_object_fid(child));
if (rc)
RETURN(rc);
}
RETURN(0);
}
int mdt_pack_encctx_in_reply(struct mdt_thread_info *info,
struct mdt_object *child)
{
struct lu_buf *buffer;
struct mdt_body *repbody;
struct req_capsule *pill = info->mti_pill;
struct obd_export *exp = mdt_info_req(info)->rq_export;
int rc = 0;
if (!exp_connect_encrypt(exp))
return rc;
if (req_capsule_has_field(pill, &RMF_FILE_ENCCTX, RCL_SERVER) &&
req_capsule_get_size(pill, &RMF_FILE_ENCCTX, RCL_SERVER) != 0) {
struct lu_attr la = { 0 };
struct dt_object *dt = mdt_obj2dt(child);
if (dt && dt->do_ops && dt->do_ops->do_attr_get)
dt_attr_get(info->mti_env, mdt_obj2dt(child), &la);
if (la.la_valid & LA_FLAGS && la.la_flags & LUSTRE_ENCRYPT_FL) {
buffer = &info->mti_buf;
/* fill reply buffer with encryption context now */
buffer->lb_len =
req_capsule_get_size(pill, &RMF_FILE_ENCCTX,
RCL_SERVER);
buffer->lb_buf =
req_capsule_server_get(pill, &RMF_FILE_ENCCTX);
rc = mo_xattr_get(info->mti_env,
mdt_object_child(child),
buffer,
LL_XATTR_NAME_ENCRYPTION_CONTEXT);
if (unlikely(rc == -ENODATA))
/* For compatibility with 2.14 */
rc = mo_xattr_get(info->mti_env,
mdt_object_child(child),
buffer,
LL_XATTR_NAME_ENCRYPTION_CONTEXT_OLD);
if (rc >= 0) {
CDEBUG(D_SEC,
"found encryption ctx of size %d for "DFID"\n",
rc, PFID(mdt_object_fid(child)));
repbody = req_capsule_server_get(pill,
&RMF_MDT_BODY);
repbody->mbo_valid |= OBD_MD_ENCCTX;
if (rc < buffer->lb_len)
req_capsule_shrink(pill,
&RMF_FILE_ENCCTX, rc,
RCL_SERVER);
rc = 0;
} else {
CDEBUG(D_SEC,
"encryption ctx not found for "DFID": rc = %d\n",
PFID(mdt_object_fid(child)), rc);
req_capsule_shrink(pill, &RMF_FILE_ENCCTX, 0,
RCL_SERVER);
/* handling -ENOENT is important because it may
* change object state in DNE env dropping
* LOHA_EXISTS flag, it is important to return
* that to the caller.
* Check LU-13115 for details.
*/
if (rc != -ENOENT)
rc = 0;
}
} else {
req_capsule_shrink(pill, &RMF_FILE_ENCCTX, 0,
RCL_SERVER);
}
}
return rc;
}
/*
* return true for a same version and false instead
*/
bool mdt_layout_version_check(struct mdt_thread_info *info,
struct mdt_object *parent, __u32 version)
{
struct md_layout_change mlc;
/* Interoperability check: skip for old clients as they don't
* provide a layout version (it is 0). Act like before.
*/
if (version == 0)
return true;
mlc.mlc_opc = MD_LAYOUT_VERSION;
mlc.mlc_layout_ver = version;
if (mo_layout_check(info->mti_env, mdt_object_child(parent), &mlc))
return false;
return true;
}
