*/
/*
* This file is part of Lustre, http://www.lustre.org/
- * Lustre is a trademark of Sun Microsystems, Inc.
*
* lustre/llite/dir.c
*
#include <lustre_fid.h>
#include <lustre_kernelcomm.h>
#include <lustre_swab.h>
+#include <libcfs/libcfs_crypto.h>
#include "llite_internal.h"
*
*/
struct page *ll_get_dir_page(struct inode *dir, struct md_op_data *op_data,
- __u64 offset)
+ __u64 offset, int *partial_readdir_rc)
{
- struct md_callback cb_op;
- struct page *page;
- int rc;
+ struct md_readdir_info mrinfo = {
+ .mr_blocking_ast = ll_md_blocking_ast };
+ struct page *page;
+ int rc;
- cb_op.md_blocking_ast = ll_md_blocking_ast;
- rc = md_read_page(ll_i2mdexp(dir), op_data, &cb_op, offset, &page);
+ rc = md_read_page(ll_i2mdexp(dir), op_data, &mrinfo, offset, &page);
if (rc != 0)
return ERR_PTR(rc);
+ if (partial_readdir_rc && mrinfo.mr_partial_readdir_rc)
+ *partial_readdir_rc = mrinfo.mr_partial_readdir_rc;
+
return page;
}
#ifdef HAVE_DIR_CONTEXT
int ll_dir_read(struct inode *inode, __u64 *ppos, struct md_op_data *op_data,
- struct dir_context *ctx)
+ struct dir_context *ctx, int *partial_readdir_rc)
{
#else
int ll_dir_read(struct inode *inode, __u64 *ppos, struct md_op_data *op_data,
- void *cookie, filldir_t filldir)
+ void *cookie, filldir_t filldir, int *partial_readdir_rc)
{
#endif
- struct ll_sb_info *sbi = ll_i2sbi(inode);
- __u64 pos = *ppos;
- bool is_api32 = ll_need_32bit_api(sbi);
- bool is_hash64 = sbi->ll_flags & LL_SBI_64BIT_HASH;
- struct page *page;
- bool done = false;
- int rc = 0;
+ struct ll_sb_info *sbi = ll_i2sbi(inode);
+ __u64 pos = *ppos;
+ bool is_api32 = ll_need_32bit_api(sbi);
+ bool is_hash64 = test_bit(LL_SBI_64BIT_HASH, sbi->ll_flags);
+ struct page *page;
+ bool done = false;
+ struct llcrypt_str lltr = LLTR_INIT(NULL, 0);
+ int rc = 0;
ENTRY;
- page = ll_get_dir_page(inode, op_data, pos);
+ if (IS_ENCRYPTED(inode)) {
+ rc = llcrypt_fname_alloc_buffer(inode, NAME_MAX, &lltr);
+ if (rc < 0)
+ RETURN(rc);
+ }
+
+ page = ll_get_dir_page(inode, op_data, pos, partial_readdir_rc);
while (rc == 0 && !done) {
struct lu_dirpage *dp;
lhash = hash;
fid_le_to_cpu(&fid, &ent->lde_fid);
ino = cl_fid_build_ino(&fid, is_api32);
- type = IFTODT(lu_dirent_type_get(ent));
+ type = S_DT(lu_dirent_type_get(ent));
/* For ll_nfs_get_name_filldir(), it will try to access
* 'ent' through 'lde_name', so the parameter 'name'
* for 'filldir()' must be part of the 'ent'. */
#ifdef HAVE_DIR_CONTEXT
ctx->pos = lhash;
- done = !dir_emit(ctx, ent->lde_name, namelen, ino,
- type);
+ if (!IS_ENCRYPTED(inode)) {
+ done = !dir_emit(ctx, ent->lde_name, namelen,
+ ino, type);
+ } else {
+ /* Directory is encrypted */
+ int save_len = lltr.len;
+ struct llcrypt_str de_name =
+ LLTR_INIT(ent->lde_name, namelen);
+
+ rc = ll_fname_disk_to_usr(inode, 0, 0, &de_name,
+ &lltr, &fid);
+ de_name = lltr;
+ lltr.len = save_len;
+ if (rc) {
+ done = 1;
+ break;
+ }
+ done = !dir_emit(ctx, de_name.name, de_name.len,
+ ino, type);
+ }
#else
+ /* HAVE_DIR_CONTEXT is defined from kernel 3.11, whereas
+ * IS_ENCRYPTED is brought by kernel 4.14.
+ * So there is no need to handle encryption case here.
+ */
done = filldir(cookie, ent->lde_name, namelen, lhash,
ino, type);
#endif
le32_to_cpu(dp->ldp_flags) &
LDF_COLLIDE);
next = pos;
- page = ll_get_dir_page(inode, op_data, pos);
+ page = ll_get_dir_page(inode, op_data, pos,
+ partial_readdir_rc);
}
}
#ifdef HAVE_DIR_CONTEXT
#else
*ppos = pos;
#endif
+ llcrypt_fname_free_buffer(&lltr);
RETURN(rc);
}
struct inode *inode = file_inode(filp);
struct ll_file_data *lfd = filp->private_data;
struct ll_sb_info *sbi = ll_i2sbi(inode);
- int hash64 = sbi->ll_flags & LL_SBI_64BIT_HASH;
+ bool hash64 = test_bit(LL_SBI_64BIT_HASH, sbi->ll_flags);
int api32 = ll_need_32bit_api(sbi);
struct md_op_data *op_data;
struct lu_fid pfid = { 0 };
ktime_t kstart = ktime_get();
+ /* result of possible partial readdir */
+ int partial_readdir_rc = 0;
__u64 pos;
int rc;
+
ENTRY;
- if (lfd != NULL)
- pos = lfd->lfd_pos;
- else
- pos = 0;
+ LASSERT(lfd != NULL);
+ pos = lfd->lfd_pos;
CDEBUG(D_VFSTRACE,
"VFS Op:inode="DFID"(%p) pos/size%lu/%llu 32bit_api %d\n",
PFID(ll_inode2fid(inode)),
inode, (unsigned long)pos, i_size_read(inode), api32);
+ if (IS_ENCRYPTED(inode)) {
+ rc = llcrypt_get_encryption_info(inode);
+ if (rc && rc != -ENOKEY)
+ GOTO(out, rc);
+ }
+
if (pos == MDS_DIR_END_OFF)
/*
* end-of-file.
#ifdef HAVE_DIR_CONTEXT
ctx->pos = pos;
- rc = ll_dir_read(inode, &pos, op_data, ctx);
+ rc = ll_dir_read(inode, &pos, op_data, ctx, &partial_readdir_rc);
pos = ctx->pos;
#else
- rc = ll_dir_read(inode, &pos, op_data, cookie, filldir);
+ rc = ll_dir_read(inode, &pos, op_data, cookie, filldir,
+ &partial_readdir_rc);
#endif
- if (lfd != NULL)
- lfd->lfd_pos = pos;
+ lfd->lfd_pos = pos;
+ if (!lfd->fd_partial_readdir_rc)
+ lfd->fd_partial_readdir_rc = partial_readdir_rc;
if (pos == MDS_DIR_END_OFF) {
if (api32)
* <0 if the creation is failed.
*/
static int ll_dir_setdirstripe(struct dentry *dparent, struct lmv_user_md *lump,
- size_t len, const char *dirname, umode_t mode)
+ size_t len, const char *dirname, umode_t mode,
+ bool createonly)
{
struct inode *parent = dparent->d_inode;
struct ptlrpc_request *request = NULL;
},
};
bool encrypt = false;
+ int hash_flags;
int err;
- ENTRY;
+ ENTRY;
if (unlikely(!lmv_user_magic_supported(lump->lum_magic)))
RETURN(-EINVAL);
!OBD_FAIL_CHECK(OBD_FAIL_LLITE_NO_CHECK_DEAD))
RETURN(-ENOENT);
+ /* MDS < 2.14 doesn't support 'crush' hash type, and cannot handle
+ * unknown hash if client doesn't set a valid one. switch to fnv_1a_64.
+ */
if (!(exp_connect_flags2(sbi->ll_md_exp) & OBD_CONNECT2_CRUSH)) {
- if ((lump->lum_hash_type & LMV_HASH_TYPE_MASK) ==
- LMV_HASH_TYPE_CRUSH) {
- /* if server doesn't support 'crush' hash type,
- * switch to fnv_1a_64.
- */
- lump->lum_hash_type &= ~LMV_HASH_TYPE_MASK;
- lump->lum_hash_type |= LMV_HASH_TYPE_FNV_1A_64;
- } else if ((lump->lum_hash_type & LMV_HASH_TYPE_MASK) ==
- LMV_HASH_TYPE_UNKNOWN) {
- /* from 2.14 MDT will choose default hash type if client
- * doesn't set a valid one, while old server doesn't
- * handle it.
- */
- lump->lum_hash_type &= ~LMV_HASH_TYPE_MASK;
- lump->lum_hash_type |= LMV_HASH_TYPE_DEFAULT;
- }
+ enum lmv_hash_type type = lump->lum_hash_type &
+ LMV_HASH_TYPE_MASK;
+
+ if (type == LMV_HASH_TYPE_CRUSH ||
+ type == LMV_HASH_TYPE_UNKNOWN)
+ lump->lum_hash_type = (lump->lum_hash_type ^ type) |
+ LMV_HASH_TYPE_FNV_1A_64;
}
+ hash_flags = lump->lum_hash_type & ~LMV_HASH_TYPE_MASK;
+ if (hash_flags & ~LMV_HASH_FLAG_KNOWN)
+ RETURN(-EINVAL);
+
if (unlikely(!lmv_user_magic_supported(cpu_to_le32(lump->lum_magic))))
lustre_swab_lmv_user_md(lump);
if (IS_ERR(op_data))
RETURN(PTR_ERR(op_data));
+ op_data->op_dir_depth = ll_i2info(parent)->lli_dir_depth;
+
if (ll_sbi_has_encrypt(sbi) &&
(IS_ENCRYPTED(parent) ||
unlikely(llcrypt_dummy_context_enabled(parent)))) {
encrypt = true;
}
- if (sbi->ll_flags & LL_SBI_FILE_SECCTX) {
+ if (test_bit(LL_SBI_FILE_SECCTX, sbi->ll_flags)) {
/* selinux_dentry_init_security() uses dentry->d_parent and name
* to determine the security context for the file. So our fake
* dentry should be real enough for this purpose. */
}
op_data->op_cli_flags |= CLI_SET_MEA;
+ if (createonly)
+ op_data->op_bias |= MDS_SETSTRIPE_CREATE;
+
err = md_create(sbi->ll_md_exp, op_data, lump, len, mode,
from_kuid(&init_user_ns, current_fsuid()),
from_kgid(&init_user_ns, current_fsgid()),
- cfs_curproc_cap_pack(), 0, &request);
+ current_cap(), 0, &request);
if (err)
GOTO(out_request, err);
CFS_FAIL_TIMEOUT(OBD_FAIL_LLITE_SETDIRSTRIPE_PAUSE, cfs_fail_val);
- err = ll_prep_inode(&inode, request, parent->i_sb, NULL);
+ err = ll_prep_inode(&inode, &request->rq_pill, parent->i_sb, NULL);
if (err)
GOTO(out_inode, err);
dentry.d_inode = inode;
- if (sbi->ll_flags & LL_SBI_FILE_SECCTX) {
+ if (test_bit(LL_SBI_FILE_SECCTX, sbi->ll_flags)) {
/* no need to protect selinux_inode_setsecurity() by
* inode_lock. Taking it would lead to a client deadlock
* LU-13617
return 0;
}
-int quotactl_ioctl(struct ll_sb_info *sbi, struct if_quotactl *qctl)
+int quotactl_ioctl(struct super_block *sb, struct if_quotactl *qctl)
{
+ struct ll_sb_info *sbi = ll_s2sbi(sb);
int cmd = qctl->qc_cmd;
int type = qctl->qc_type;
int id = qctl->qc_id;
case LUSTRE_Q_SETDEFAULT:
case LUSTRE_Q_SETQUOTAPOOL:
case LUSTRE_Q_SETINFOPOOL:
- if (!cfs_capable(CAP_SYS_ADMIN))
+ case LUSTRE_Q_SETDEFAULT_POOL:
+ if (!capable(CAP_SYS_ADMIN))
RETURN(-EPERM);
+
+ if (sb->s_flags & SB_RDONLY)
+ RETURN(-EROFS);
break;
case Q_GETQUOTA:
case LUSTRE_Q_GETDEFAULT:
case LUSTRE_Q_GETQUOTAPOOL:
+ case LUSTRE_Q_GETDEFAULT_POOL:
if (check_owner(type, id) &&
- (!cfs_capable(CAP_SYS_ADMIN)))
+ (!capable(CAP_SYS_ADMIN)))
RETURN(-EPERM);
break;
case Q_GETINFO:
int ll_rmfid(struct file *file, void __user *arg)
{
const struct fid_array __user *ufa = arg;
+ struct inode *inode = file_inode(file);
struct fid_array *lfa = NULL;
size_t size;
unsigned nr;
int i, rc, *rcs = NULL;
ENTRY;
- if (!cfs_capable(CAP_DAC_READ_SEARCH) &&
- !(ll_i2sbi(file_inode(file))->ll_flags & LL_SBI_USER_FID2PATH))
+ if (!capable(CAP_DAC_READ_SEARCH) &&
+ !test_bit(LL_SBI_USER_FID2PATH, ll_i2sbi(inode)->ll_flags))
RETURN(-EPERM);
/* Only need to get the buflen */
if (get_user(nr, &ufa->fa_nr))
}
case LL_IOC_LMV_SETSTRIPE: {
struct lmv_user_md *lum;
- char *filename;
- int namelen = 0;
- int lumlen = 0;
- umode_t mode;
- int len;
- int rc;
+ char *filename;
+ int namelen = 0;
+ int lumlen = 0;
+ umode_t mode;
+ bool createonly = false;
+ int len;
+ int rc;
rc = obd_ioctl_getdata(&data, &len, (void __user *)arg);
if (rc)
}
mode = data->ioc_type;
- rc = ll_dir_setdirstripe(dentry, lum, lumlen, filename, mode);
+ createonly = data->ioc_obdo1.o_flags & OBD_FL_OBDMDEXISTS;
+ rc = ll_dir_setdirstripe(dentry, lum, lumlen, filename, mode,
+ createonly);
lmv_out_free:
OBD_FREE_LARGE(data, len);
RETURN(rc);
__u32 __user *lmmsizep = NULL;
struct lu_fid __user *fidp = NULL;
int lmmsize;
+ bool api32;
if (cmd == IOC_MDC_GETFILEINFO_V1 ||
cmd == IOC_MDC_GETFILEINFO_V2 ||
GOTO(out_req, rc = -EFAULT);
rc = -EOVERFLOW;
}
+ api32 = test_bit(LL_SBI_32BIT_API, sbi->ll_flags);
if (cmd == IOC_MDC_GETFILEINFO_V1 ||
cmd == LL_IOC_MDC_GETINFO_V1) {
st.st_uid = body->mbo_uid;
st.st_gid = body->mbo_gid;
st.st_rdev = body->mbo_rdev;
- st.st_size = body->mbo_size;
+ if (llcrypt_require_key(inode) == -ENOKEY)
+ st.st_size = round_up(st.st_size,
+ LUSTRE_ENCRYPTION_UNIT_SIZE);
+ else
+ st.st_size = body->mbo_size;
st.st_blksize = PAGE_SIZE;
st.st_blocks = body->mbo_blocks;
st.st_atime = body->mbo_atime;
st.st_mtime = body->mbo_mtime;
st.st_ctime = body->mbo_ctime;
st.st_ino = cl_fid_build_ino(&body->mbo_fid1,
- sbi->ll_flags &
- LL_SBI_32BIT_API);
+ api32);
if (copy_to_user(statp, &st, sizeof(st)))
GOTO(out_req, rc = -EFAULT);
stx.stx_gid = body->mbo_gid;
stx.stx_mode = body->mbo_mode;
stx.stx_ino = cl_fid_build_ino(&body->mbo_fid1,
- sbi->ll_flags &
- LL_SBI_32BIT_API);
- stx.stx_size = body->mbo_size;
+ api32);
+ if (llcrypt_require_key(inode) == -ENOKEY)
+ stx.stx_size = round_up(stx.stx_size,
+ LUSTRE_ENCRYPTION_UNIT_SIZE);
+ else
+ stx.stx_size = body->mbo_size;
stx.stx_blocks = body->mbo_blocks;
stx.stx_atime.tv_sec = body->mbo_atime;
stx.stx_ctime.tv_sec = body->mbo_ctime;
GOTO(out_quotactl, rc = -EFAULT);
}
- rc = quotactl_ioctl(sbi, qctl);
+ rc = quotactl_ioctl(inode->i_sb, qctl);
if (rc == 0 &&
copy_to_user((void __user *)arg, qctl, sizeof(*qctl)))
rc = -EFAULT;
OBD_FREE(qctl, qctl_len);
RETURN(rc);
}
- case OBD_IOC_GETDTNAME:
- case OBD_IOC_GETMDNAME:
+ case OBD_IOC_GETNAME_OLD:
+ /* fall through */
+ case OBD_IOC_GETDTNAME:
+ /* fall through */
+ case OBD_IOC_GETMDNAME:
RETURN(ll_get_obd_name(inode, cmd, arg));
case LL_IOC_FLUSHCTX:
RETURN(ll_flush_ctx(inode));
RETURN(rc);
}
case LL_IOC_HSM_CT_START:
- if (!cfs_capable(CAP_SYS_ADMIN))
+ if (!capable(CAP_SYS_ADMIN))
RETURN(-EPERM);
rc = copy_and_ct_start(cmd, sbi->ll_md_exp,
int len;
char *filename;
int namelen = 0;
+ __u32 flags;
int rc;
rc = obd_ioctl_getdata(&data, &len, (void __user *)arg);
filename = data->ioc_inlbuf1;
namelen = data->ioc_inllen1;
+ flags = data->ioc_type;
if (namelen < 1 || namelen != strlen(filename) + 1) {
CDEBUG(D_INFO, "IOC_MDC_LOOKUP missing filename\n");
GOTO(migrate_free, rc);
}
- rc = ll_migrate(inode, file, lum, filename);
+ rc = ll_migrate(inode, file, lum, filename, flags);
migrate_free:
OBD_FREE_LARGE(data, len);
RETURN(ll_ioctl_fsgetxattr(inode, cmd, arg));
case FS_IOC_FSSETXATTR:
RETURN(ll_ioctl_fssetxattr(inode, cmd, arg));
+ case LL_IOC_PROJECT:
+ RETURN(ll_ioctl_project(file, cmd, arg));
case LL_IOC_PCC_DETACH_BY_FID: {
struct lu_pcc_detach_fid *detach;
struct lu_fid *fid;
if (!S_ISREG(inode2->i_mode))
GOTO(out_iput, rc = -EINVAL);
- if (!inode_owner_or_capable(inode2))
+ if (!inode_owner_or_capable(&init_user_ns, inode2))
GOTO(out_iput, rc = -EPERM);
rc = pcc_ioctl_detach(inode2, detach->pccd_opt);
((api32 && offset <= LL_DIR_END_OFF_32BIT) ||
(!api32 && offset <= LL_DIR_END_OFF))) {
if (offset != file->f_pos) {
+ bool hash64;
+
+ hash64 = test_bit(LL_SBI_64BIT_HASH, sbi->ll_flags);
if ((api32 && offset == LL_DIR_END_OFF_32BIT) ||
(!api32 && offset == LL_DIR_END_OFF))
fd->lfd_pos = MDS_DIR_END_OFF;
- else if (api32 && sbi->ll_flags & LL_SBI_64BIT_HASH)
+ else if (api32 && hash64)
fd->lfd_pos = offset << 32;
else
fd->lfd_pos = offset;
RETURN(ll_file_release(inode, file));
}
+/* notify error if partially read striped directory */
+static int ll_dir_flush(struct file *file, fl_owner_t id)
+{
+ struct ll_file_data *lfd = file->private_data;
+ int rc = lfd->fd_partial_readdir_rc;
+
+ lfd->fd_partial_readdir_rc = 0;
+
+ return rc;
+}
+
const struct file_operations ll_dir_operations = {
.llseek = ll_dir_seek,
.open = ll_dir_open,
#endif
.unlocked_ioctl = ll_dir_ioctl,
.fsync = ll_fsync,
+ .flush = ll_dir_flush,
};
git://git.whamcloud.com / fs / lustre-release.git / blobdiff