2015-07-17 14:38:14 +00:00
|
|
|
/*
|
|
|
|
* (C) 2001 Clemson University and The University of Chicago
|
|
|
|
*
|
|
|
|
* See COPYING in top-level directory.
|
|
|
|
*/
|
|
|
|
#include "protocol.h"
|
2015-12-04 17:56:14 +00:00
|
|
|
#include "orangefs-kernel.h"
|
|
|
|
#include "orangefs-dev-proto.h"
|
|
|
|
#include "orangefs-bufmap.h"
|
2015-07-17 14:38:14 +00:00
|
|
|
|
2015-11-24 20:12:14 +00:00
|
|
|
__s32 fsid_of_op(struct orangefs_kernel_op_s *op)
|
2015-07-17 14:38:14 +00:00
|
|
|
{
|
2015-11-24 20:12:14 +00:00
|
|
|
__s32 fsid = ORANGEFS_FS_ID_NULL;
|
2015-07-17 14:38:14 +00:00
|
|
|
|
|
|
|
if (op) {
|
|
|
|
switch (op->upcall.type) {
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_FILE_IO:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.io.refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_LOOKUP:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.lookup.parent_refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_CREATE:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.create.parent_refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_GETATTR:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.getattr.refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_REMOVE:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.remove.parent_refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_MKDIR:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.mkdir.parent_refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_READDIR:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.readdir.refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_SETATTR:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.setattr.refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_SYMLINK:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.sym.parent_refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_RENAME:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.rename.old_parent_refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_STATFS:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.statfs.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_TRUNCATE:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.truncate.refn.fs_id;
|
|
|
|
break;
|
2016-08-02 18:31:05 +00:00
|
|
|
case ORANGEFS_VFS_OP_RA_FLUSH:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.ra_cache_flush.refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_FS_UMOUNT:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.fs_umount.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_GETXATTR:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.getxattr.refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_SETXATTR:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.setxattr.refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_LISTXATTR:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.listxattr.refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_REMOVEXATTR:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.removexattr.refn.fs_id;
|
|
|
|
break;
|
2015-11-24 20:12:14 +00:00
|
|
|
case ORANGEFS_VFS_OP_FSYNC:
|
2015-07-17 14:38:14 +00:00
|
|
|
fsid = op->upcall.req.fsync.refn.fs_id;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return fsid;
|
|
|
|
}
|
|
|
|
|
2016-01-25 20:33:39 +00:00
|
|
|
static int orangefs_inode_flags(struct ORANGEFS_sys_attr_s *attrs)
|
2015-07-17 14:38:14 +00:00
|
|
|
{
|
2016-01-25 20:33:39 +00:00
|
|
|
int flags = 0;
|
2015-11-24 20:12:14 +00:00
|
|
|
if (attrs->flags & ORANGEFS_IMMUTABLE_FL)
|
2016-01-25 20:33:39 +00:00
|
|
|
flags |= S_IMMUTABLE;
|
2015-07-17 14:38:14 +00:00
|
|
|
else
|
2016-01-25 20:33:39 +00:00
|
|
|
flags &= ~S_IMMUTABLE;
|
2015-11-24 20:12:14 +00:00
|
|
|
if (attrs->flags & ORANGEFS_APPEND_FL)
|
2016-01-25 20:33:39 +00:00
|
|
|
flags |= S_APPEND;
|
2015-07-17 14:38:14 +00:00
|
|
|
else
|
2016-01-25 20:33:39 +00:00
|
|
|
flags &= ~S_APPEND;
|
2015-11-24 20:12:14 +00:00
|
|
|
if (attrs->flags & ORANGEFS_NOATIME_FL)
|
2016-01-25 20:33:39 +00:00
|
|
|
flags |= S_NOATIME;
|
2015-07-17 14:38:14 +00:00
|
|
|
else
|
2016-01-25 20:33:39 +00:00
|
|
|
flags &= ~S_NOATIME;
|
|
|
|
return flags;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int orangefs_inode_perms(struct ORANGEFS_sys_attr_s *attrs)
|
|
|
|
{
|
|
|
|
int perm_mode = 0;
|
|
|
|
|
|
|
|
if (attrs->perms & ORANGEFS_O_EXECUTE)
|
|
|
|
perm_mode |= S_IXOTH;
|
|
|
|
if (attrs->perms & ORANGEFS_O_WRITE)
|
|
|
|
perm_mode |= S_IWOTH;
|
|
|
|
if (attrs->perms & ORANGEFS_O_READ)
|
|
|
|
perm_mode |= S_IROTH;
|
|
|
|
|
|
|
|
if (attrs->perms & ORANGEFS_G_EXECUTE)
|
|
|
|
perm_mode |= S_IXGRP;
|
|
|
|
if (attrs->perms & ORANGEFS_G_WRITE)
|
|
|
|
perm_mode |= S_IWGRP;
|
|
|
|
if (attrs->perms & ORANGEFS_G_READ)
|
|
|
|
perm_mode |= S_IRGRP;
|
2015-07-17 14:38:14 +00:00
|
|
|
|
2016-01-25 20:33:39 +00:00
|
|
|
if (attrs->perms & ORANGEFS_U_EXECUTE)
|
|
|
|
perm_mode |= S_IXUSR;
|
|
|
|
if (attrs->perms & ORANGEFS_U_WRITE)
|
|
|
|
perm_mode |= S_IWUSR;
|
|
|
|
if (attrs->perms & ORANGEFS_U_READ)
|
|
|
|
perm_mode |= S_IRUSR;
|
|
|
|
|
|
|
|
if (attrs->perms & ORANGEFS_G_SGID)
|
|
|
|
perm_mode |= S_ISGID;
|
|
|
|
if (attrs->perms & ORANGEFS_U_SUID)
|
|
|
|
perm_mode |= S_ISUID;
|
|
|
|
|
|
|
|
return perm_mode;
|
2015-07-17 14:38:14 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* NOTE: in kernel land, we never use the sys_attr->link_target for
|
|
|
|
* anything, so don't bother copying it into the sys_attr object here.
|
|
|
|
*/
|
|
|
|
static inline int copy_attributes_from_inode(struct inode *inode,
|
2015-11-24 20:12:14 +00:00
|
|
|
struct ORANGEFS_sys_attr_s *attrs,
|
2015-07-17 14:38:14 +00:00
|
|
|
struct iattr *iattr)
|
|
|
|
{
|
|
|
|
umode_t tmp_mode;
|
|
|
|
|
|
|
|
if (!iattr || !inode || !attrs) {
|
|
|
|
gossip_err("NULL iattr (%p), inode (%p), attrs (%p) "
|
|
|
|
"in copy_attributes_from_inode!\n",
|
|
|
|
iattr,
|
|
|
|
inode,
|
|
|
|
attrs);
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
/*
|
|
|
|
* We need to be careful to only copy the attributes out of the
|
|
|
|
* iattr object that we know are valid.
|
|
|
|
*/
|
|
|
|
attrs->mask = 0;
|
|
|
|
if (iattr->ia_valid & ATTR_UID) {
|
orangefs: fix namespace handling
In orangefs_inode_getxattr(), an fsuid is written to dmesg. The kuid is
converted to a userspace uid via from_kuid(current_user_ns(), [...]), but
since dmesg is global, init_user_ns should be used here instead.
In copy_attributes_from_inode(), op_alloc() and fill_default_sys_attrs(),
upcall structures are populated with uids/gids that have been mapped into
the caller's namespace. However, those upcall structures are read by
another process (the userspace filesystem driver), and that process might
be running in another namespace. This effectively lets any user spoof its
uid and gid as seen by the userspace filesystem driver.
To fix the second issue, I just construct the opcall structures with
init_user_ns uids/gids and require the filesystem server to run in the
init namespace. Since orangefs is full of global state anyway (as the error
message in DUMP_DEVICE_ERROR explains, there can only be one userspace
orangefs filesystem driver at once), that shouldn't be a problem.
[
Why does orangefs even exist in the kernel if everything does upcalls into
userspace? What does orangefs do that couldn't be done with the FUSE
interface? If there is no good answer to those questions, I'd prefer to see
orangefs kicked out of the kernel. Can that be done for something that
shipped in a release?
According to commit f7ab093f74bf ("Orangefs: kernel client part 1"), they
even already have a FUSE daemon, and the only rational reason (apart from
"but most of our users report preferring to use our kernel module instead")
given for not wanting to use FUSE is one "in-the-works" feature that could
probably be integated into FUSE instead.
]
This patch has been compile-tested.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
2016-06-24 23:51:52 +00:00
|
|
|
attrs->owner = from_kuid(&init_user_ns, iattr->ia_uid);
|
2015-11-24 20:12:14 +00:00
|
|
|
attrs->mask |= ORANGEFS_ATTR_SYS_UID;
|
2015-07-17 14:38:14 +00:00
|
|
|
gossip_debug(GOSSIP_UTILS_DEBUG, "(UID) %d\n", attrs->owner);
|
|
|
|
}
|
|
|
|
if (iattr->ia_valid & ATTR_GID) {
|
orangefs: fix namespace handling
In orangefs_inode_getxattr(), an fsuid is written to dmesg. The kuid is
converted to a userspace uid via from_kuid(current_user_ns(), [...]), but
since dmesg is global, init_user_ns should be used here instead.
In copy_attributes_from_inode(), op_alloc() and fill_default_sys_attrs(),
upcall structures are populated with uids/gids that have been mapped into
the caller's namespace. However, those upcall structures are read by
another process (the userspace filesystem driver), and that process might
be running in another namespace. This effectively lets any user spoof its
uid and gid as seen by the userspace filesystem driver.
To fix the second issue, I just construct the opcall structures with
init_user_ns uids/gids and require the filesystem server to run in the
init namespace. Since orangefs is full of global state anyway (as the error
message in DUMP_DEVICE_ERROR explains, there can only be one userspace
orangefs filesystem driver at once), that shouldn't be a problem.
[
Why does orangefs even exist in the kernel if everything does upcalls into
userspace? What does orangefs do that couldn't be done with the FUSE
interface? If there is no good answer to those questions, I'd prefer to see
orangefs kicked out of the kernel. Can that be done for something that
shipped in a release?
According to commit f7ab093f74bf ("Orangefs: kernel client part 1"), they
even already have a FUSE daemon, and the only rational reason (apart from
"but most of our users report preferring to use our kernel module instead")
given for not wanting to use FUSE is one "in-the-works" feature that could
probably be integated into FUSE instead.
]
This patch has been compile-tested.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
2016-06-24 23:51:52 +00:00
|
|
|
attrs->group = from_kgid(&init_user_ns, iattr->ia_gid);
|
2015-11-24 20:12:14 +00:00
|
|
|
attrs->mask |= ORANGEFS_ATTR_SYS_GID;
|
2015-07-17 14:38:14 +00:00
|
|
|
gossip_debug(GOSSIP_UTILS_DEBUG, "(GID) %d\n", attrs->group);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (iattr->ia_valid & ATTR_ATIME) {
|
2015-11-24 20:12:14 +00:00
|
|
|
attrs->mask |= ORANGEFS_ATTR_SYS_ATIME;
|
2015-07-17 14:38:14 +00:00
|
|
|
if (iattr->ia_valid & ATTR_ATIME_SET) {
|
2016-02-26 12:54:10 +00:00
|
|
|
attrs->atime = (time64_t)iattr->ia_atime.tv_sec;
|
2015-11-24 20:12:14 +00:00
|
|
|
attrs->mask |= ORANGEFS_ATTR_SYS_ATIME_SET;
|
2015-07-17 14:38:14 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
if (iattr->ia_valid & ATTR_MTIME) {
|
2015-11-24 20:12:14 +00:00
|
|
|
attrs->mask |= ORANGEFS_ATTR_SYS_MTIME;
|
2015-07-17 14:38:14 +00:00
|
|
|
if (iattr->ia_valid & ATTR_MTIME_SET) {
|
2016-02-26 12:54:10 +00:00
|
|
|
attrs->mtime = (time64_t)iattr->ia_mtime.tv_sec;
|
2015-11-24 20:12:14 +00:00
|
|
|
attrs->mask |= ORANGEFS_ATTR_SYS_MTIME_SET;
|
2015-07-17 14:38:14 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
if (iattr->ia_valid & ATTR_CTIME)
|
2015-11-24 20:12:14 +00:00
|
|
|
attrs->mask |= ORANGEFS_ATTR_SYS_CTIME;
|
2015-07-17 14:38:14 +00:00
|
|
|
|
|
|
|
/*
|
2015-11-24 20:12:14 +00:00
|
|
|
* ORANGEFS cannot set size with a setattr operation. Probably not likely
|
2015-07-17 14:38:14 +00:00
|
|
|
* to be requested through the VFS, but just in case, don't worry about
|
|
|
|
* ATTR_SIZE
|
|
|
|
*/
|
|
|
|
|
|
|
|
if (iattr->ia_valid & ATTR_MODE) {
|
|
|
|
tmp_mode = iattr->ia_mode;
|
|
|
|
if (tmp_mode & (S_ISVTX)) {
|
|
|
|
if (is_root_handle(inode)) {
|
|
|
|
/*
|
|
|
|
* allow sticky bit to be set on root (since
|
|
|
|
* it shows up that way by default anyhow),
|
|
|
|
* but don't show it to the server
|
|
|
|
*/
|
|
|
|
tmp_mode -= S_ISVTX;
|
|
|
|
} else {
|
|
|
|
gossip_debug(GOSSIP_UTILS_DEBUG,
|
|
|
|
"User attempted to set sticky bit on non-root directory; returning EINVAL.\n");
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (tmp_mode & (S_ISUID)) {
|
|
|
|
gossip_debug(GOSSIP_UTILS_DEBUG,
|
|
|
|
"Attempting to set setuid bit (not supported); returning EINVAL.\n");
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
2015-11-24 20:12:14 +00:00
|
|
|
attrs->perms = ORANGEFS_util_translate_mode(tmp_mode);
|
|
|
|
attrs->mask |= ORANGEFS_ATTR_SYS_PERM;
|
2015-07-17 14:38:14 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2016-03-15 15:28:20 +00:00
|
|
|
static int orangefs_inode_type(enum orangefs_ds_type objtype)
|
|
|
|
{
|
|
|
|
if (objtype == ORANGEFS_TYPE_METAFILE)
|
|
|
|
return S_IFREG;
|
|
|
|
else if (objtype == ORANGEFS_TYPE_DIRECTORY)
|
|
|
|
return S_IFDIR;
|
|
|
|
else if (objtype == ORANGEFS_TYPE_SYMLINK)
|
|
|
|
return S_IFLNK;
|
|
|
|
else
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2016-03-17 20:01:52 +00:00
|
|
|
static int orangefs_inode_is_stale(struct inode *inode, int new,
|
|
|
|
struct ORANGEFS_sys_attr_s *attrs, char *link_target)
|
|
|
|
{
|
|
|
|
struct orangefs_inode_s *orangefs_inode = ORANGEFS_I(inode);
|
|
|
|
int type = orangefs_inode_type(attrs->objtype);
|
|
|
|
if (!new) {
|
|
|
|
/*
|
|
|
|
* If the inode type or symlink target have changed then this
|
|
|
|
* inode is stale.
|
|
|
|
*/
|
|
|
|
if (type == -1 || !(inode->i_mode & type)) {
|
|
|
|
orangefs_make_bad_inode(inode);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
if (type == S_IFLNK && strncmp(orangefs_inode->link_target,
|
|
|
|
link_target, ORANGEFS_NAME_MAX)) {
|
|
|
|
orangefs_make_bad_inode(inode);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2017-04-25 19:38:03 +00:00
|
|
|
int orangefs_inode_getattr(struct inode *inode, int new, int bypass,
|
|
|
|
u32 request_mask)
|
2016-03-15 15:28:20 +00:00
|
|
|
{
|
|
|
|
struct orangefs_inode_s *orangefs_inode = ORANGEFS_I(inode);
|
|
|
|
struct orangefs_kernel_op_s *new_op;
|
|
|
|
loff_t inode_size, rounded_up_size;
|
2016-03-17 20:01:52 +00:00
|
|
|
int ret, type;
|
2016-03-15 15:28:20 +00:00
|
|
|
|
|
|
|
gossip_debug(GOSSIP_UTILS_DEBUG, "%s: called on inode %pU\n", __func__,
|
|
|
|
get_khandle_from_ino(inode));
|
|
|
|
|
2016-06-09 20:32:38 +00:00
|
|
|
if (!new && !bypass) {
|
2017-04-25 19:38:03 +00:00
|
|
|
/*
|
|
|
|
* Must have all the attributes in the mask and be within cache
|
|
|
|
* time.
|
|
|
|
*/
|
|
|
|
if ((request_mask & orangefs_inode->getattr_mask) ==
|
|
|
|
request_mask &&
|
|
|
|
time_before(jiffies, orangefs_inode->getattr_time))
|
2016-06-09 20:32:38 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2016-03-15 15:28:20 +00:00
|
|
|
new_op = op_alloc(ORANGEFS_VFS_OP_GETATTR);
|
|
|
|
if (!new_op)
|
|
|
|
return -ENOMEM;
|
|
|
|
new_op->upcall.req.getattr.refn = orangefs_inode->refn;
|
2017-04-25 19:38:03 +00:00
|
|
|
/*
|
|
|
|
* Size is the hardest attribute to get. The incremental cost of any
|
|
|
|
* other attribute is essentially zero.
|
|
|
|
*/
|
|
|
|
if (request_mask & STATX_SIZE || new)
|
|
|
|
new_op->upcall.req.getattr.mask = ORANGEFS_ATTR_SYS_ALL_NOHINT;
|
|
|
|
else
|
|
|
|
new_op->upcall.req.getattr.mask =
|
|
|
|
ORANGEFS_ATTR_SYS_ALL_NOHINT & ~ORANGEFS_ATTR_SYS_SIZE;
|
2016-03-15 15:28:20 +00:00
|
|
|
|
|
|
|
ret = service_operation(new_op, __func__,
|
|
|
|
get_interruptible_flag(inode));
|
|
|
|
if (ret != 0)
|
|
|
|
goto out;
|
|
|
|
|
2016-03-17 20:01:52 +00:00
|
|
|
type = orangefs_inode_type(new_op->
|
2016-03-15 15:28:20 +00:00
|
|
|
downcall.resp.getattr.attributes.objtype);
|
2016-03-17 20:01:52 +00:00
|
|
|
ret = orangefs_inode_is_stale(inode, new,
|
|
|
|
&new_op->downcall.resp.getattr.attributes,
|
|
|
|
new_op->downcall.resp.getattr.link_target);
|
|
|
|
if (ret) {
|
|
|
|
ret = -ESTALE;
|
|
|
|
goto out;
|
2016-03-15 15:28:20 +00:00
|
|
|
}
|
|
|
|
|
2016-03-17 20:01:52 +00:00
|
|
|
switch (type) {
|
2016-03-15 15:28:20 +00:00
|
|
|
case S_IFREG:
|
|
|
|
inode->i_flags = orangefs_inode_flags(&new_op->
|
|
|
|
downcall.resp.getattr.attributes);
|
2017-04-25 19:38:03 +00:00
|
|
|
if (request_mask & STATX_SIZE || new) {
|
|
|
|
inode_size = (loff_t)new_op->
|
|
|
|
downcall.resp.getattr.attributes.size;
|
|
|
|
rounded_up_size =
|
|
|
|
(inode_size + (4096 - (inode_size % 4096)));
|
|
|
|
inode->i_size = inode_size;
|
|
|
|
orangefs_inode->blksize =
|
|
|
|
new_op->downcall.resp.getattr.attributes.blksize;
|
|
|
|
spin_lock(&inode->i_lock);
|
|
|
|
inode->i_bytes = inode_size;
|
|
|
|
inode->i_blocks =
|
|
|
|
(unsigned long)(rounded_up_size / 512);
|
|
|
|
spin_unlock(&inode->i_lock);
|
|
|
|
}
|
2016-03-15 15:28:20 +00:00
|
|
|
break;
|
|
|
|
case S_IFDIR:
|
2017-04-25 19:38:03 +00:00
|
|
|
if (request_mask & STATX_SIZE || new) {
|
|
|
|
inode->i_size = PAGE_SIZE;
|
|
|
|
orangefs_inode->blksize = i_blocksize(inode);
|
|
|
|
spin_lock(&inode->i_lock);
|
|
|
|
inode_set_bytes(inode, inode->i_size);
|
|
|
|
spin_unlock(&inode->i_lock);
|
|
|
|
}
|
2016-03-15 15:28:20 +00:00
|
|
|
set_nlink(inode, 1);
|
|
|
|
break;
|
|
|
|
case S_IFLNK:
|
|
|
|
if (new) {
|
|
|
|
inode->i_size = (loff_t)strlen(new_op->
|
|
|
|
downcall.resp.getattr.link_target);
|
2017-02-27 22:28:32 +00:00
|
|
|
orangefs_inode->blksize = i_blocksize(inode);
|
2016-04-08 17:33:21 +00:00
|
|
|
ret = strscpy(orangefs_inode->link_target,
|
2016-03-15 15:28:20 +00:00
|
|
|
new_op->downcall.resp.getattr.link_target,
|
|
|
|
ORANGEFS_NAME_MAX);
|
2016-04-08 17:33:21 +00:00
|
|
|
if (ret == -E2BIG) {
|
|
|
|
ret = -EIO;
|
|
|
|
goto out;
|
|
|
|
}
|
2016-03-18 18:20:15 +00:00
|
|
|
inode->i_link = orangefs_inode->link_target;
|
2016-03-15 15:28:20 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
inode->i_uid = make_kuid(&init_user_ns, new_op->
|
|
|
|
downcall.resp.getattr.attributes.owner);
|
|
|
|
inode->i_gid = make_kgid(&init_user_ns, new_op->
|
|
|
|
downcall.resp.getattr.attributes.group);
|
|
|
|
inode->i_atime.tv_sec = (time64_t)new_op->
|
|
|
|
downcall.resp.getattr.attributes.atime;
|
|
|
|
inode->i_mtime.tv_sec = (time64_t)new_op->
|
|
|
|
downcall.resp.getattr.attributes.mtime;
|
|
|
|
inode->i_ctime.tv_sec = (time64_t)new_op->
|
|
|
|
downcall.resp.getattr.attributes.ctime;
|
|
|
|
inode->i_atime.tv_nsec = 0;
|
|
|
|
inode->i_mtime.tv_nsec = 0;
|
|
|
|
inode->i_ctime.tv_nsec = 0;
|
|
|
|
|
|
|
|
/* special case: mark the root inode as sticky */
|
2016-03-17 20:01:52 +00:00
|
|
|
inode->i_mode = type | (is_root_handle(inode) ? S_ISVTX : 0) |
|
2016-03-15 15:28:20 +00:00
|
|
|
orangefs_inode_perms(&new_op->downcall.resp.getattr.attributes);
|
|
|
|
|
2016-08-16 15:38:14 +00:00
|
|
|
orangefs_inode->getattr_time = jiffies +
|
|
|
|
orangefs_getattr_timeout_msecs*HZ/1000;
|
2017-04-25 19:38:03 +00:00
|
|
|
if (request_mask & STATX_SIZE || new)
|
|
|
|
orangefs_inode->getattr_mask = STATX_BASIC_STATS;
|
|
|
|
else
|
|
|
|
orangefs_inode->getattr_mask = STATX_BASIC_STATS & ~STATX_SIZE;
|
2016-03-15 15:28:20 +00:00
|
|
|
ret = 0;
|
|
|
|
out:
|
|
|
|
op_release(new_op);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2016-03-17 19:15:16 +00:00
|
|
|
int orangefs_inode_check_changed(struct inode *inode)
|
|
|
|
{
|
|
|
|
struct orangefs_inode_s *orangefs_inode = ORANGEFS_I(inode);
|
|
|
|
struct orangefs_kernel_op_s *new_op;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
gossip_debug(GOSSIP_UTILS_DEBUG, "%s: called on inode %pU\n", __func__,
|
|
|
|
get_khandle_from_ino(inode));
|
|
|
|
|
|
|
|
new_op = op_alloc(ORANGEFS_VFS_OP_GETATTR);
|
|
|
|
if (!new_op)
|
|
|
|
return -ENOMEM;
|
|
|
|
new_op->upcall.req.getattr.refn = orangefs_inode->refn;
|
|
|
|
new_op->upcall.req.getattr.mask = ORANGEFS_ATTR_SYS_TYPE |
|
|
|
|
ORANGEFS_ATTR_SYS_LNK_TARGET;
|
|
|
|
|
|
|
|
ret = service_operation(new_op, __func__,
|
|
|
|
get_interruptible_flag(inode));
|
|
|
|
if (ret != 0)
|
|
|
|
goto out;
|
|
|
|
|
2016-03-17 20:01:52 +00:00
|
|
|
ret = orangefs_inode_is_stale(inode, 0,
|
|
|
|
&new_op->downcall.resp.getattr.attributes,
|
|
|
|
new_op->downcall.resp.getattr.link_target);
|
2016-03-17 19:15:16 +00:00
|
|
|
out:
|
|
|
|
op_release(new_op);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2015-07-17 14:38:14 +00:00
|
|
|
/*
|
2015-11-24 20:12:14 +00:00
|
|
|
* issues a orangefs setattr request to make sure the new attribute values
|
2015-07-17 14:38:14 +00:00
|
|
|
* take effect if successful. returns 0 on success; -errno otherwise
|
|
|
|
*/
|
2015-11-24 20:12:14 +00:00
|
|
|
int orangefs_inode_setattr(struct inode *inode, struct iattr *iattr)
|
2015-07-17 14:38:14 +00:00
|
|
|
{
|
2015-11-24 20:12:14 +00:00
|
|
|
struct orangefs_inode_s *orangefs_inode = ORANGEFS_I(inode);
|
|
|
|
struct orangefs_kernel_op_s *new_op;
|
2015-07-17 14:38:14 +00:00
|
|
|
int ret;
|
|
|
|
|
2015-11-24 20:12:14 +00:00
|
|
|
new_op = op_alloc(ORANGEFS_VFS_OP_SETATTR);
|
2015-07-17 14:38:14 +00:00
|
|
|
if (!new_op)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
2015-11-24 20:12:14 +00:00
|
|
|
new_op->upcall.req.setattr.refn = orangefs_inode->refn;
|
2015-07-17 14:38:14 +00:00
|
|
|
ret = copy_attributes_from_inode(inode,
|
|
|
|
&new_op->upcall.req.setattr.attributes,
|
|
|
|
iattr);
|
2016-01-23 00:47:47 +00:00
|
|
|
if (ret >= 0) {
|
|
|
|
ret = service_operation(new_op, __func__,
|
2015-07-17 14:38:14 +00:00
|
|
|
get_interruptible_flag(inode));
|
|
|
|
|
2016-01-23 00:47:47 +00:00
|
|
|
gossip_debug(GOSSIP_UTILS_DEBUG,
|
|
|
|
"orangefs_inode_setattr: returning %d\n",
|
|
|
|
ret);
|
|
|
|
}
|
2015-07-17 14:38:14 +00:00
|
|
|
|
|
|
|
op_release(new_op);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* successful setattr should clear the atime, mtime and
|
|
|
|
* ctime flags.
|
|
|
|
*/
|
|
|
|
if (ret == 0) {
|
2015-11-24 20:12:14 +00:00
|
|
|
ClearAtimeFlag(orangefs_inode);
|
|
|
|
ClearMtimeFlag(orangefs_inode);
|
|
|
|
ClearCtimeFlag(orangefs_inode);
|
|
|
|
ClearModeFlag(orangefs_inode);
|
2016-07-28 18:46:36 +00:00
|
|
|
orangefs_inode->getattr_time = jiffies - 1;
|
2015-07-17 14:38:14 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2015-11-24 20:12:14 +00:00
|
|
|
int orangefs_flush_inode(struct inode *inode)
|
2015-07-17 14:38:14 +00:00
|
|
|
{
|
|
|
|
/*
|
|
|
|
* If it is a dirty inode, this function gets called.
|
|
|
|
* Gather all the information that needs to be setattr'ed
|
|
|
|
* Right now, this will only be used for mode, atime, mtime
|
|
|
|
* and/or ctime.
|
|
|
|
*/
|
|
|
|
struct iattr wbattr;
|
|
|
|
int ret;
|
|
|
|
int mtime_flag;
|
|
|
|
int ctime_flag;
|
|
|
|
int atime_flag;
|
|
|
|
int mode_flag;
|
2015-11-24 20:12:14 +00:00
|
|
|
struct orangefs_inode_s *orangefs_inode = ORANGEFS_I(inode);
|
2015-07-17 14:38:14 +00:00
|
|
|
|
|
|
|
memset(&wbattr, 0, sizeof(wbattr));
|
|
|
|
|
|
|
|
/*
|
|
|
|
* check inode flags up front, and clear them if they are set. This
|
|
|
|
* will prevent multiple processes from all trying to flush the same
|
|
|
|
* inode if they call close() simultaneously
|
|
|
|
*/
|
2015-11-24 20:12:14 +00:00
|
|
|
mtime_flag = MtimeFlag(orangefs_inode);
|
|
|
|
ClearMtimeFlag(orangefs_inode);
|
|
|
|
ctime_flag = CtimeFlag(orangefs_inode);
|
|
|
|
ClearCtimeFlag(orangefs_inode);
|
|
|
|
atime_flag = AtimeFlag(orangefs_inode);
|
|
|
|
ClearAtimeFlag(orangefs_inode);
|
|
|
|
mode_flag = ModeFlag(orangefs_inode);
|
|
|
|
ClearModeFlag(orangefs_inode);
|
2015-07-17 14:38:14 +00:00
|
|
|
|
|
|
|
/* -- Lazy atime,mtime and ctime update --
|
|
|
|
* Note: all times are dictated by server in the new scheme
|
|
|
|
* and not by the clients
|
|
|
|
*
|
|
|
|
* Also mode updates are being handled now..
|
|
|
|
*/
|
|
|
|
|
|
|
|
if (mtime_flag)
|
|
|
|
wbattr.ia_valid |= ATTR_MTIME;
|
|
|
|
if (ctime_flag)
|
|
|
|
wbattr.ia_valid |= ATTR_CTIME;
|
|
|
|
if (atime_flag)
|
|
|
|
wbattr.ia_valid |= ATTR_ATIME;
|
|
|
|
|
|
|
|
if (mode_flag) {
|
|
|
|
wbattr.ia_mode = inode->i_mode;
|
|
|
|
wbattr.ia_valid |= ATTR_MODE;
|
|
|
|
}
|
|
|
|
|
|
|
|
gossip_debug(GOSSIP_UTILS_DEBUG,
|
2015-11-24 20:12:14 +00:00
|
|
|
"*********** orangefs_flush_inode: %pU "
|
2015-07-17 14:38:14 +00:00
|
|
|
"(ia_valid %d)\n",
|
|
|
|
get_khandle_from_ino(inode),
|
|
|
|
wbattr.ia_valid);
|
|
|
|
if (wbattr.ia_valid == 0) {
|
|
|
|
gossip_debug(GOSSIP_UTILS_DEBUG,
|
2015-11-24 20:12:14 +00:00
|
|
|
"orangefs_flush_inode skipping setattr()\n");
|
2015-07-17 14:38:14 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
gossip_debug(GOSSIP_UTILS_DEBUG,
|
2015-11-24 20:12:14 +00:00
|
|
|
"orangefs_flush_inode (%pU) writing mode %o\n",
|
2015-07-17 14:38:14 +00:00
|
|
|
get_khandle_from_ino(inode),
|
|
|
|
inode->i_mode);
|
|
|
|
|
2015-11-24 20:12:14 +00:00
|
|
|
ret = orangefs_inode_setattr(inode, &wbattr);
|
2015-07-17 14:38:14 +00:00
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2015-11-24 20:12:14 +00:00
|
|
|
void orangefs_make_bad_inode(struct inode *inode)
|
2015-07-17 14:38:14 +00:00
|
|
|
{
|
|
|
|
if (is_root_handle(inode)) {
|
|
|
|
/*
|
|
|
|
* if this occurs, the pvfs2-client-core was killed but we
|
|
|
|
* can't afford to lose the inode operations and such
|
|
|
|
* associated with the root handle in any case.
|
|
|
|
*/
|
|
|
|
gossip_debug(GOSSIP_UTILS_DEBUG,
|
|
|
|
"*** NOT making bad root inode %pU\n",
|
|
|
|
get_khandle_from_ino(inode));
|
|
|
|
} else {
|
|
|
|
gossip_debug(GOSSIP_UTILS_DEBUG,
|
|
|
|
"*** making bad inode %pU\n",
|
|
|
|
get_khandle_from_ino(inode));
|
|
|
|
make_bad_inode(inode);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2015-10-05 17:44:24 +00:00
|
|
|
/*
|
|
|
|
* The following is a very dirty hack that is now a permanent part of the
|
2015-11-24 20:12:14 +00:00
|
|
|
* ORANGEFS protocol. See protocol.h for more error definitions.
|
2015-10-05 17:44:24 +00:00
|
|
|
*/
|
2015-10-02 16:11:19 +00:00
|
|
|
|
2015-11-24 20:12:14 +00:00
|
|
|
/* The order matches include/orangefs-types.h in the OrangeFS source. */
|
2015-10-02 16:11:19 +00:00
|
|
|
static int PINT_errno_mapping[] = {
|
2015-10-05 17:44:24 +00:00
|
|
|
0, EPERM, ENOENT, EINTR, EIO, ENXIO, EBADF, EAGAIN, ENOMEM,
|
|
|
|
EFAULT, EBUSY, EEXIST, ENODEV, ENOTDIR, EISDIR, EINVAL, EMFILE,
|
|
|
|
EFBIG, ENOSPC, EROFS, EMLINK, EPIPE, EDEADLK, ENAMETOOLONG,
|
|
|
|
ENOLCK, ENOSYS, ENOTEMPTY, ELOOP, EWOULDBLOCK, ENOMSG, EUNATCH,
|
|
|
|
EBADR, EDEADLOCK, ENODATA, ETIME, ENONET, EREMOTE, ECOMM,
|
|
|
|
EPROTO, EBADMSG, EOVERFLOW, ERESTART, EMSGSIZE, EPROTOTYPE,
|
|
|
|
ENOPROTOOPT, EPROTONOSUPPORT, EOPNOTSUPP, EADDRINUSE,
|
|
|
|
EADDRNOTAVAIL, ENETDOWN, ENETUNREACH, ENETRESET, ENOBUFS,
|
|
|
|
ETIMEDOUT, ECONNREFUSED, EHOSTDOWN, EHOSTUNREACH, EALREADY,
|
|
|
|
EACCES, ECONNRESET, ERANGE
|
2015-10-02 16:11:19 +00:00
|
|
|
};
|
2015-07-17 14:38:14 +00:00
|
|
|
|
2015-11-24 20:12:14 +00:00
|
|
|
int orangefs_normalize_to_errno(__s32 error_code)
|
2015-07-17 14:38:14 +00:00
|
|
|
{
|
2015-10-05 17:44:24 +00:00
|
|
|
__u32 i;
|
|
|
|
|
2015-10-02 16:11:19 +00:00
|
|
|
/* Success */
|
|
|
|
if (error_code == 0) {
|
|
|
|
return 0;
|
2015-10-05 17:44:24 +00:00
|
|
|
/*
|
|
|
|
* This shouldn't ever happen. If it does it should be fixed on the
|
|
|
|
* server.
|
|
|
|
*/
|
2015-10-02 16:11:19 +00:00
|
|
|
} else if (error_code > 0) {
|
2015-11-24 20:12:14 +00:00
|
|
|
gossip_err("orangefs: error status receieved.\n");
|
|
|
|
gossip_err("orangefs: assuming error code is inverted.\n");
|
2015-07-17 14:38:14 +00:00
|
|
|
error_code = -error_code;
|
|
|
|
}
|
|
|
|
|
2015-10-05 17:44:24 +00:00
|
|
|
/*
|
2015-11-24 20:12:14 +00:00
|
|
|
* XXX: This is very bad since error codes from ORANGEFS may not be
|
2015-10-05 17:44:24 +00:00
|
|
|
* suitable for return into userspace.
|
|
|
|
*/
|
2015-10-02 16:11:19 +00:00
|
|
|
|
2015-10-05 17:44:24 +00:00
|
|
|
/*
|
2015-11-24 20:12:14 +00:00
|
|
|
* Convert ORANGEFS error values into errno values suitable for return
|
2015-10-05 17:44:24 +00:00
|
|
|
* from the kernel.
|
|
|
|
*/
|
2015-11-24 20:12:14 +00:00
|
|
|
if ((-error_code) & ORANGEFS_NON_ERRNO_ERROR_BIT) {
|
2015-10-02 16:11:19 +00:00
|
|
|
if (((-error_code) &
|
2015-11-24 20:12:14 +00:00
|
|
|
(ORANGEFS_ERROR_NUMBER_BITS|ORANGEFS_NON_ERRNO_ERROR_BIT|
|
|
|
|
ORANGEFS_ERROR_BIT)) == ORANGEFS_ECANCEL) {
|
2015-07-17 14:38:14 +00:00
|
|
|
/*
|
|
|
|
* cancellation error codes generally correspond to
|
|
|
|
* a timeout from the client's perspective
|
|
|
|
*/
|
|
|
|
error_code = -ETIMEDOUT;
|
|
|
|
} else {
|
|
|
|
/* assume a default error code */
|
2015-11-24 20:12:14 +00:00
|
|
|
gossip_err("orangefs: warning: got error code without errno equivalent: %d.\n", error_code);
|
2015-07-17 14:38:14 +00:00
|
|
|
error_code = -EINVAL;
|
|
|
|
}
|
2015-10-02 16:11:19 +00:00
|
|
|
|
2015-11-24 20:12:14 +00:00
|
|
|
/* Convert ORANGEFS encoded errno values into regular errno values. */
|
|
|
|
} else if ((-error_code) & ORANGEFS_ERROR_BIT) {
|
|
|
|
i = (-error_code) & ~(ORANGEFS_ERROR_BIT|ORANGEFS_ERROR_CLASS_BITS);
|
2015-10-05 17:44:24 +00:00
|
|
|
if (i < sizeof(PINT_errno_mapping)/sizeof(*PINT_errno_mapping))
|
2015-10-02 16:11:19 +00:00
|
|
|
error_code = -PINT_errno_mapping[i];
|
|
|
|
else
|
|
|
|
error_code = -EINVAL;
|
|
|
|
|
2015-10-05 17:44:24 +00:00
|
|
|
/*
|
2015-11-24 20:12:14 +00:00
|
|
|
* Only ORANGEFS protocol error codes should ever come here. Otherwise
|
2015-10-05 17:44:24 +00:00
|
|
|
* there is a bug somewhere.
|
|
|
|
*/
|
2015-10-02 16:11:19 +00:00
|
|
|
} else {
|
2015-11-24 20:12:14 +00:00
|
|
|
gossip_err("orangefs: orangefs_normalize_to_errno: got error code which is not from ORANGEFS.\n");
|
2015-07-17 14:38:14 +00:00
|
|
|
}
|
|
|
|
return error_code;
|
|
|
|
}
|
|
|
|
|
|
|
|
#define NUM_MODES 11
|
2015-11-24 20:12:14 +00:00
|
|
|
__s32 ORANGEFS_util_translate_mode(int mode)
|
2015-07-17 14:38:14 +00:00
|
|
|
{
|
|
|
|
int ret = 0;
|
|
|
|
int i = 0;
|
|
|
|
static int modes[NUM_MODES] = {
|
|
|
|
S_IXOTH, S_IWOTH, S_IROTH,
|
|
|
|
S_IXGRP, S_IWGRP, S_IRGRP,
|
|
|
|
S_IXUSR, S_IWUSR, S_IRUSR,
|
|
|
|
S_ISGID, S_ISUID
|
|
|
|
};
|
2015-11-24 20:12:14 +00:00
|
|
|
static int orangefs_modes[NUM_MODES] = {
|
|
|
|
ORANGEFS_O_EXECUTE, ORANGEFS_O_WRITE, ORANGEFS_O_READ,
|
|
|
|
ORANGEFS_G_EXECUTE, ORANGEFS_G_WRITE, ORANGEFS_G_READ,
|
|
|
|
ORANGEFS_U_EXECUTE, ORANGEFS_U_WRITE, ORANGEFS_U_READ,
|
|
|
|
ORANGEFS_G_SGID, ORANGEFS_U_SUID
|
2015-07-17 14:38:14 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
for (i = 0; i < NUM_MODES; i++)
|
|
|
|
if (mode & modes[i])
|
2015-11-24 20:12:14 +00:00
|
|
|
ret |= orangefs_modes[i];
|
2015-07-17 14:38:14 +00:00
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
#undef NUM_MODES
|