diff options
Diffstat (limited to 'fs/xfs/xfs_ioctl.c')
| -rw-r--r-- | fs/xfs/xfs_ioctl.c | 2214 |
1 files changed, 1002 insertions, 1212 deletions
diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 6e2bca5d44d6..59eaad774371 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -1,950 +1,748 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Copyright (c) 2000-2005 Silicon Graphics, Inc. * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" #include "xfs_fs.h" -#include "xfs_log.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_alloc.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_log_format.h" +#include "xfs_trans_resv.h" #include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_dinode.h" #include "xfs_inode.h" -#include "xfs_ioctl.h" #include "xfs_rtalloc.h" +#include "xfs_iwalk.h" #include "xfs_itable.h" #include "xfs_error.h" +#include "xfs_da_format.h" +#include "xfs_da_btree.h" #include "xfs_attr.h" #include "xfs_bmap.h" -#include "xfs_buf_item.h" -#include "xfs_utils.h" -#include "xfs_dfrag.h" +#include "xfs_bmap_util.h" #include "xfs_fsops.h" -#include "xfs_vnodeops.h" #include "xfs_discard.h" #include "xfs_quota.h" -#include "xfs_inode_item.h" -#include "xfs_export.h" #include "xfs_trace.h" #include "xfs_icache.h" +#include "xfs_trans.h" +#include "xfs_btree.h" +#include <linux/fsmap.h> +#include "xfs_fsmap.h" +#include "scrub/xfs_scrub.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_health.h" +#include "xfs_reflink.h" +#include "xfs_ioctl.h" +#include "xfs_xattr.h" +#include "xfs_rtbitmap.h" +#include "xfs_file.h" +#include "xfs_exchrange.h" +#include "xfs_handle.h" +#include "xfs_rtgroup.h" -#include <linux/capability.h> -#include <linux/dcache.h> #include <linux/mount.h> -#include <linux/namei.h> -#include <linux/pagemap.h> -#include <linux/slab.h> -#include <linux/exportfs.h> +#include <linux/fileattr.h> + +/* Return 0 on success or positive error */ +int +xfs_fsbulkstat_one_fmt( + struct xfs_ibulk *breq, + const struct xfs_bulkstat *bstat) +{ + struct xfs_bstat bs1; + + xfs_bulkstat_to_bstat(breq->mp, &bs1, bstat); + if (copy_to_user(breq->ubuffer, &bs1, sizeof(bs1))) + return -EFAULT; + return xfs_ibulk_advance(breq, sizeof(struct xfs_bstat)); +} -/* - * xfs_find_handle maps from userspace xfs_fsop_handlereq structure to - * a file or fs handle. - * - * XFS_IOC_PATH_TO_FSHANDLE - * returns fs handle for a mount point or path within that mount point - * XFS_IOC_FD_TO_HANDLE - * returns full handle for a FD opened in user space - * XFS_IOC_PATH_TO_HANDLE - * returns full handle for a path - */ int -xfs_find_handle( +xfs_fsinumbers_fmt( + struct xfs_ibulk *breq, + const struct xfs_inumbers *igrp) +{ + struct xfs_inogrp ig1; + + xfs_inumbers_to_inogrp(&ig1, igrp); + if (copy_to_user(breq->ubuffer, &ig1, sizeof(struct xfs_inogrp))) + return -EFAULT; + return xfs_ibulk_advance(breq, sizeof(struct xfs_inogrp)); +} + +STATIC int +xfs_ioc_fsbulkstat( + struct file *file, unsigned int cmd, - xfs_fsop_handlereq_t *hreq) + void __user *arg) { - int hsize; - xfs_handle_t handle; - struct inode *inode; - struct fd f = {0}; - struct path path; + struct xfs_mount *mp = XFS_I(file_inode(file))->i_mount; + struct xfs_fsop_bulkreq bulkreq; + struct xfs_ibulk breq = { + .mp = mp, + .idmap = file_mnt_idmap(file), + .ocount = 0, + }; + xfs_ino_t lastino; int error; - struct xfs_inode *ip; - - if (cmd == XFS_IOC_FD_TO_HANDLE) { - f = fdget(hreq->fd); - if (!f.file) - return -EBADF; - inode = file_inode(f.file); - } else { - error = user_lpath((const char __user *)hreq->path, &path); - if (error) - return error; - inode = path.dentry->d_inode; - } - ip = XFS_I(inode); - /* - * We can only generate handles for inodes residing on a XFS filesystem, - * and only for regular files, directories or symbolic links. - */ - error = -EINVAL; - if (inode->i_sb->s_magic != XFS_SB_MAGIC) - goto out_put; + /* done = 1 if there are more stats to get and if bulkstat */ + /* should be called again (unused here, but used in dmapi) */ - error = -EBADF; - if (!S_ISREG(inode->i_mode) && - !S_ISDIR(inode->i_mode) && - !S_ISLNK(inode->i_mode)) - goto out_put; + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + if (xfs_is_shutdown(mp)) + return -EIO; - memcpy(&handle.ha_fsid, ip->i_mount->m_fixedfsid, sizeof(xfs_fsid_t)); + if (copy_from_user(&bulkreq, arg, sizeof(struct xfs_fsop_bulkreq))) + return -EFAULT; - if (cmd == XFS_IOC_PATH_TO_FSHANDLE) { - /* - * This handle only contains an fsid, zero the rest. - */ - memset(&handle.ha_fid, 0, sizeof(handle.ha_fid)); - hsize = sizeof(xfs_fsid_t); - } else { - int lock_mode; - - lock_mode = xfs_ilock_map_shared(ip); - handle.ha_fid.fid_len = sizeof(xfs_fid_t) - - sizeof(handle.ha_fid.fid_len); - handle.ha_fid.fid_pad = 0; - handle.ha_fid.fid_gen = ip->i_d.di_gen; - handle.ha_fid.fid_ino = ip->i_ino; - xfs_iunlock_map_shared(ip, lock_mode); - - hsize = XFS_HSIZE(handle); + if (copy_from_user(&lastino, bulkreq.lastip, sizeof(__s64))) + return -EFAULT; + + if (bulkreq.icount <= 0) + return -EINVAL; + + if (bulkreq.ubuffer == NULL) + return -EINVAL; + + breq.ubuffer = bulkreq.ubuffer; + breq.icount = bulkreq.icount; + + /* + * FSBULKSTAT_SINGLE expects that *lastip contains the inode number + * that we want to stat. However, FSINUMBERS and FSBULKSTAT expect + * that *lastip contains either zero or the number of the last inode to + * be examined by the previous call and return results starting with + * the next inode after that. The new bulk request back end functions + * take the inode to start with, so we have to compute the startino + * parameter from lastino to maintain correct function. lastino == 0 + * is a special case because it has traditionally meant "first inode + * in filesystem". + */ + if (cmd == XFS_IOC_FSINUMBERS) { + breq.startino = lastino ? lastino + 1 : 0; + error = xfs_inumbers(&breq, xfs_fsinumbers_fmt); + lastino = breq.startino - 1; + } else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE) { + breq.startino = lastino; + breq.icount = 1; + error = xfs_bulkstat_one(&breq, xfs_fsbulkstat_one_fmt); + } else { /* XFS_IOC_FSBULKSTAT */ + breq.startino = lastino ? lastino + 1 : 0; + error = xfs_bulkstat(&breq, xfs_fsbulkstat_one_fmt); + lastino = breq.startino - 1; } - error = -EFAULT; - if (copy_to_user(hreq->ohandle, &handle, hsize) || - copy_to_user(hreq->ohandlen, &hsize, sizeof(__s32))) - goto out_put; + if (error) + return error; - error = 0; + if (bulkreq.lastip != NULL && + copy_to_user(bulkreq.lastip, &lastino, sizeof(xfs_ino_t))) + return -EFAULT; - out_put: - if (cmd == XFS_IOC_FD_TO_HANDLE) - fdput(f); - else - path_put(&path); - return error; + if (bulkreq.ocount != NULL && + copy_to_user(bulkreq.ocount, &breq.ocount, sizeof(__s32))) + return -EFAULT; + + return 0; } -/* - * No need to do permission checks on the various pathname components - * as the handle operations are privileged. - */ -STATIC int -xfs_handle_acceptable( - void *context, - struct dentry *dentry) +/* Return 0 on success or positive error */ +static int +xfs_bulkstat_fmt( + struct xfs_ibulk *breq, + const struct xfs_bulkstat *bstat) { - return 1; + if (copy_to_user(breq->ubuffer, bstat, sizeof(struct xfs_bulkstat))) + return -EFAULT; + return xfs_ibulk_advance(breq, sizeof(struct xfs_bulkstat)); } /* - * Convert userspace handle data into a dentry. + * Check the incoming bulk request @hdr from userspace and initialize the + * internal @breq bulk request appropriately. Returns 0 if the bulk request + * should proceed; -ECANCELED if there's nothing to do; or the usual + * negative error code. */ -struct dentry * -xfs_handle_to_dentry( - struct file *parfilp, - void __user *uhandle, - u32 hlen) +static int +xfs_bulk_ireq_setup( + struct xfs_mount *mp, + const struct xfs_bulk_ireq *hdr, + struct xfs_ibulk *breq, + void __user *ubuffer) { - xfs_handle_t handle; - struct xfs_fid64 fid; + if (hdr->icount == 0 || + (hdr->flags & ~XFS_BULK_IREQ_FLAGS_ALL) || + memchr_inv(hdr->reserved, 0, sizeof(hdr->reserved))) + return -EINVAL; + + breq->startino = hdr->ino; + breq->ubuffer = ubuffer; + breq->icount = hdr->icount; + breq->ocount = 0; + breq->flags = 0; /* - * Only allow handle opens under a directory. + * The @ino parameter is a special value, so we must look it up here. + * We're not allowed to have IREQ_AGNO, and we only return one inode + * worth of data. */ - if (!S_ISDIR(file_inode(parfilp)->i_mode)) - return ERR_PTR(-ENOTDIR); - - if (hlen != sizeof(xfs_handle_t)) - return ERR_PTR(-EINVAL); - if (copy_from_user(&handle, uhandle, hlen)) - return ERR_PTR(-EFAULT); - if (handle.ha_fid.fid_len != - sizeof(handle.ha_fid) - sizeof(handle.ha_fid.fid_len)) - return ERR_PTR(-EINVAL); - - memset(&fid, 0, sizeof(struct fid)); - fid.ino = handle.ha_fid.fid_ino; - fid.gen = handle.ha_fid.fid_gen; - - return exportfs_decode_fh(parfilp->f_path.mnt, (struct fid *)&fid, 3, - FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG, - xfs_handle_acceptable, NULL); -} - -STATIC struct dentry * -xfs_handlereq_to_dentry( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq) -{ - return xfs_handle_to_dentry(parfilp, hreq->ihandle, hreq->ihandlen); -} - -int -xfs_open_by_handle( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq) -{ - const struct cred *cred = current_cred(); - int error; - int fd; - int permflag; - struct file *filp; - struct inode *inode; - struct dentry *dentry; - fmode_t fmode; - struct path path; - - if (!capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - - dentry = xfs_handlereq_to_dentry(parfilp, hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - inode = dentry->d_inode; + if (hdr->flags & XFS_BULK_IREQ_SPECIAL) { + if (hdr->flags & XFS_BULK_IREQ_AGNO) + return -EINVAL; - /* Restrict xfs_open_by_handle to directories & regular files. */ - if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode))) { - error = -XFS_ERROR(EPERM); - goto out_dput; + switch (hdr->ino) { + case XFS_BULK_IREQ_SPECIAL_ROOT: + breq->startino = mp->m_sb.sb_rootino; + break; + default: + return -EINVAL; + } + breq->icount = 1; } -#if BITS_PER_LONG != 32 - hreq->oflags |= O_LARGEFILE; -#endif + /* + * The IREQ_AGNO flag means that we only want results from a given AG. + * If @hdr->ino is zero, we start iterating in that AG. If @hdr->ino is + * beyond the specified AG then we return no results. + */ + if (hdr->flags & XFS_BULK_IREQ_AGNO) { + if (hdr->agno >= mp->m_sb.sb_agcount) + return -EINVAL; - permflag = hreq->oflags; - fmode = OPEN_FMODE(permflag); - if ((!(permflag & O_APPEND) || (permflag & O_TRUNC)) && - (fmode & FMODE_WRITE) && IS_APPEND(inode)) { - error = -XFS_ERROR(EPERM); - goto out_dput; - } + if (breq->startino == 0) + breq->startino = XFS_AGINO_TO_INO(mp, hdr->agno, 0); + else if (XFS_INO_TO_AGNO(mp, breq->startino) < hdr->agno) + return -EINVAL; - if ((fmode & FMODE_WRITE) && IS_IMMUTABLE(inode)) { - error = -XFS_ERROR(EACCES); - goto out_dput; - } + breq->iwalk_flags |= XFS_IWALK_SAME_AG; - /* Can't write directories. */ - if (S_ISDIR(inode->i_mode) && (fmode & FMODE_WRITE)) { - error = -XFS_ERROR(EISDIR); - goto out_dput; - } - - fd = get_unused_fd_flags(0); - if (fd < 0) { - error = fd; - goto out_dput; - } + /* Asking for an inode past the end of the AG? We're done! */ + if (XFS_INO_TO_AGNO(mp, breq->startino) > hdr->agno) + return -ECANCELED; + } else if (hdr->agno) + return -EINVAL; - path.mnt = parfilp->f_path.mnt; - path.dentry = dentry; - filp = dentry_open(&path, hreq->oflags, cred); - dput(dentry); - if (IS_ERR(filp)) { - put_unused_fd(fd); - return PTR_ERR(filp); - } + /* Asking for an inode past the end of the FS? We're done! */ + if (XFS_INO_TO_AGNO(mp, breq->startino) >= mp->m_sb.sb_agcount) + return -ECANCELED; - if (S_ISREG(inode->i_mode)) { - filp->f_flags |= O_NOATIME; - filp->f_mode |= FMODE_NOCMTIME; - } + if (hdr->flags & XFS_BULK_IREQ_NREXT64) + breq->flags |= XFS_IBULK_NREXT64; - fd_install(fd, filp); - return fd; + /* Caller wants to see metadata directories in bulkstat output. */ + if (hdr->flags & XFS_BULK_IREQ_METADIR) + breq->flags |= XFS_IBULK_METADIR; - out_dput: - dput(dentry); - return error; + return 0; } /* - * This is a copy from fs/namei.c:vfs_readlink(), except for removing it's - * unused first argument. + * Update the userspace bulk request @hdr to reflect the end state of the + * internal bulk request @breq. */ -STATIC int -do_readlink( - char __user *buffer, - int buflen, - const char *link) +static void +xfs_bulk_ireq_teardown( + struct xfs_bulk_ireq *hdr, + struct xfs_ibulk *breq) { - int len; - - len = PTR_ERR(link); - if (IS_ERR(link)) - goto out; - - len = strlen(link); - if (len > (unsigned) buflen) - len = buflen; - if (copy_to_user(buffer, link, len)) - len = -EFAULT; - out: - return len; + hdr->ino = breq->startino; + hdr->ocount = breq->ocount; } - -int -xfs_readlink_by_handle( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq) +/* Handle the v5 bulkstat ioctl. */ +STATIC int +xfs_ioc_bulkstat( + struct file *file, + unsigned int cmd, + struct xfs_bulkstat_req __user *arg) { - struct dentry *dentry; - __u32 olen; - void *link; - int error; + struct xfs_mount *mp = XFS_I(file_inode(file))->i_mount; + struct xfs_bulk_ireq hdr; + struct xfs_ibulk breq = { + .mp = mp, + .idmap = file_mnt_idmap(file), + }; + int error; if (!capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - - dentry = xfs_handlereq_to_dentry(parfilp, hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); + return -EPERM; - /* Restrict this handle operation to symlinks only. */ - if (!S_ISLNK(dentry->d_inode->i_mode)) { - error = -XFS_ERROR(EINVAL); - goto out_dput; - } + if (xfs_is_shutdown(mp)) + return -EIO; - if (copy_from_user(&olen, hreq->ohandlen, sizeof(__u32))) { - error = -XFS_ERROR(EFAULT); - goto out_dput; - } + if (copy_from_user(&hdr, &arg->hdr, sizeof(hdr))) + return -EFAULT; - link = kmalloc(MAXPATHLEN+1, GFP_KERNEL); - if (!link) { - error = -XFS_ERROR(ENOMEM); - goto out_dput; - } + error = xfs_bulk_ireq_setup(mp, &hdr, &breq, arg->bulkstat); + if (error == -ECANCELED) + goto out_teardown; + if (error < 0) + return error; - error = -xfs_readlink(XFS_I(dentry->d_inode), link); + error = xfs_bulkstat(&breq, xfs_bulkstat_fmt); if (error) - goto out_kfree; - error = do_readlink(hreq->ohandle, olen, link); - if (error) - goto out_kfree; + return error; - out_kfree: - kfree(link); - out_dput: - dput(dentry); - return error; +out_teardown: + xfs_bulk_ireq_teardown(&hdr, &breq); + if (copy_to_user(&arg->hdr, &hdr, sizeof(hdr))) + return -EFAULT; + + return 0; } STATIC int -xfs_fssetdm_by_handle( - struct file *parfilp, - void __user *arg) +xfs_inumbers_fmt( + struct xfs_ibulk *breq, + const struct xfs_inumbers *igrp) { - int error; - struct fsdmidata fsd; - xfs_fsop_setdm_handlereq_t dmhreq; - struct dentry *dentry; - - if (!capable(CAP_MKNOD)) - return -XFS_ERROR(EPERM); - if (copy_from_user(&dmhreq, arg, sizeof(xfs_fsop_setdm_handlereq_t))) - return -XFS_ERROR(EFAULT); - - error = mnt_want_write_file(parfilp); - if (error) - return error; - - dentry = xfs_handlereq_to_dentry(parfilp, &dmhreq.hreq); - if (IS_ERR(dentry)) { - mnt_drop_write_file(parfilp); - return PTR_ERR(dentry); - } - - if (IS_IMMUTABLE(dentry->d_inode) || IS_APPEND(dentry->d_inode)) { - error = -XFS_ERROR(EPERM); - goto out; - } - - if (copy_from_user(&fsd, dmhreq.data, sizeof(fsd))) { - error = -XFS_ERROR(EFAULT); - goto out; - } - - error = -xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask, - fsd.fsd_dmstate); - - out: - mnt_drop_write_file(parfilp); - dput(dentry); - return error; + if (copy_to_user(breq->ubuffer, igrp, sizeof(struct xfs_inumbers))) + return -EFAULT; + return xfs_ibulk_advance(breq, sizeof(struct xfs_inumbers)); } +/* Handle the v5 inumbers ioctl. */ STATIC int -xfs_attrlist_by_handle( - struct file *parfilp, - void __user *arg) +xfs_ioc_inumbers( + struct xfs_mount *mp, + unsigned int cmd, + struct xfs_inumbers_req __user *arg) { - int error = -ENOMEM; - attrlist_cursor_kern_t *cursor; - xfs_fsop_attrlist_handlereq_t al_hreq; - struct dentry *dentry; - char *kbuf; + struct xfs_bulk_ireq hdr; + struct xfs_ibulk breq = { + .mp = mp, + }; + int error; if (!capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - if (copy_from_user(&al_hreq, arg, sizeof(xfs_fsop_attrlist_handlereq_t))) - return -XFS_ERROR(EFAULT); - if (al_hreq.buflen > XATTR_LIST_MAX) - return -XFS_ERROR(EINVAL); - - /* - * Reject flags, only allow namespaces. - */ - if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE)) - return -XFS_ERROR(EINVAL); - - dentry = xfs_handlereq_to_dentry(parfilp, &al_hreq.hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - - kbuf = kmem_zalloc(al_hreq.buflen, KM_SLEEP | KM_MAYFAIL); - if (!kbuf) { - kbuf = kmem_zalloc_large(al_hreq.buflen); - if (!kbuf) - goto out_dput; - } + return -EPERM; - cursor = (attrlist_cursor_kern_t *)&al_hreq.pos; - error = -xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen, - al_hreq.flags, cursor); - if (error) - goto out_kfree; + if (xfs_is_shutdown(mp)) + return -EIO; - if (copy_to_user(al_hreq.buffer, kbuf, al_hreq.buflen)) - error = -EFAULT; + if (copy_from_user(&hdr, &arg->hdr, sizeof(hdr))) + return -EFAULT; - out_kfree: - if (is_vmalloc_addr(kbuf)) - kmem_free_large(kbuf); - else - kmem_free(kbuf); - out_dput: - dput(dentry); - return error; -} + if (hdr.flags & XFS_BULK_IREQ_METADIR) + return -EINVAL; -int -xfs_attrmulti_attr_get( - struct inode *inode, - unsigned char *name, - unsigned char __user *ubuf, - __uint32_t *len, - __uint32_t flags) -{ - unsigned char *kbuf; - int error = EFAULT; - - if (*len > XATTR_SIZE_MAX) - return EINVAL; - kbuf = kmem_zalloc(*len, KM_SLEEP | KM_MAYFAIL); - if (!kbuf) { - kbuf = kmem_zalloc_large(*len); - if (!kbuf) - return ENOMEM; - } + error = xfs_bulk_ireq_setup(mp, &hdr, &breq, arg->inumbers); + if (error == -ECANCELED) + goto out_teardown; + if (error < 0) + return error; - error = xfs_attr_get(XFS_I(inode), name, kbuf, (int *)len, flags); + error = xfs_inumbers(&breq, xfs_inumbers_fmt); if (error) - goto out_kfree; + return error; - if (copy_to_user(ubuf, kbuf, *len)) - error = EFAULT; +out_teardown: + xfs_bulk_ireq_teardown(&hdr, &breq); + if (copy_to_user(&arg->hdr, &hdr, sizeof(hdr))) + return -EFAULT; - out_kfree: - if (is_vmalloc_addr(kbuf)) - kmem_free_large(kbuf); - else - kmem_free(kbuf); - return error; + return 0; } -int -xfs_attrmulti_attr_set( - struct inode *inode, - unsigned char *name, - const unsigned char __user *ubuf, - __uint32_t len, - __uint32_t flags) +STATIC int +xfs_ioc_fsgeometry( + struct xfs_mount *mp, + void __user *arg, + int struct_version) { - unsigned char *kbuf; - int error = EFAULT; - - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - return EPERM; - if (len > XATTR_SIZE_MAX) - return EINVAL; - - kbuf = memdup_user(ubuf, len); - if (IS_ERR(kbuf)) - return PTR_ERR(kbuf); - - error = xfs_attr_set(XFS_I(inode), name, kbuf, len, flags); - - return error; -} + struct xfs_fsop_geom fsgeo; + size_t len; + + xfs_fs_geometry(mp, &fsgeo, struct_version); + + if (struct_version <= 3) + len = sizeof(struct xfs_fsop_geom_v1); + else if (struct_version == 4) + len = sizeof(struct xfs_fsop_geom_v4); + else { + xfs_fsop_geom_health(mp, &fsgeo); + len = sizeof(fsgeo); + } -int -xfs_attrmulti_attr_remove( - struct inode *inode, - unsigned char *name, - __uint32_t flags) -{ - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - return EPERM; - return xfs_attr_remove(XFS_I(inode), name, flags); + if (copy_to_user(arg, &fsgeo, len)) + return -EFAULT; + return 0; } STATIC int -xfs_attrmulti_by_handle( - struct file *parfilp, +xfs_ioc_ag_geometry( + struct xfs_mount *mp, void __user *arg) { + struct xfs_perag *pag; + struct xfs_ag_geometry ageo; int error; - xfs_attr_multiop_t *ops; - xfs_fsop_attrmulti_handlereq_t am_hreq; - struct dentry *dentry; - unsigned int i, size; - unsigned char *attr_name; - - if (!capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - if (copy_from_user(&am_hreq, arg, sizeof(xfs_fsop_attrmulti_handlereq_t))) - return -XFS_ERROR(EFAULT); - - /* overflow check */ - if (am_hreq.opcount >= INT_MAX / sizeof(xfs_attr_multiop_t)) - return -E2BIG; - - dentry = xfs_handlereq_to_dentry(parfilp, &am_hreq.hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - - error = E2BIG; - size = am_hreq.opcount * sizeof(xfs_attr_multiop_t); - if (!size || size > 16 * PAGE_SIZE) - goto out_dput; - - ops = memdup_user(am_hreq.ops, size); - if (IS_ERR(ops)) { - error = PTR_ERR(ops); - goto out_dput; - } - attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL); - if (!attr_name) - goto out_kfree_ops; + if (copy_from_user(&ageo, arg, sizeof(ageo))) + return -EFAULT; + if (ageo.ag_flags) + return -EINVAL; + if (memchr_inv(&ageo.ag_reserved, 0, sizeof(ageo.ag_reserved))) + return -EINVAL; - error = 0; - for (i = 0; i < am_hreq.opcount; i++) { - ops[i].am_error = strncpy_from_user((char *)attr_name, - ops[i].am_attrname, MAXNAMELEN); - if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN) - error = -ERANGE; - if (ops[i].am_error < 0) - break; + pag = xfs_perag_get(mp, ageo.ag_number); + if (!pag) + return -EINVAL; - switch (ops[i].am_opcode) { - case ATTR_OP_GET: - ops[i].am_error = xfs_attrmulti_attr_get( - dentry->d_inode, attr_name, - ops[i].am_attrvalue, &ops[i].am_length, - ops[i].am_flags); - break; - case ATTR_OP_SET: - ops[i].am_error = mnt_want_write_file(parfilp); - if (ops[i].am_error) - break; - ops[i].am_error = xfs_attrmulti_attr_set( - dentry->d_inode, attr_name, - ops[i].am_attrvalue, ops[i].am_length, - ops[i].am_flags); - mnt_drop_write_file(parfilp); - break; - case ATTR_OP_REMOVE: - ops[i].am_error = mnt_want_write_file(parfilp); - if (ops[i].am_error) - break; - ops[i].am_error = xfs_attrmulti_attr_remove( - dentry->d_inode, attr_name, - ops[i].am_flags); - mnt_drop_write_file(parfilp); - break; - default: - ops[i].am_error = EINVAL; - } - } - - if (copy_to_user(am_hreq.ops, ops, size)) - error = XFS_ERROR(EFAULT); + error = xfs_ag_get_geometry(pag, &ageo); + xfs_perag_put(pag); + if (error) + return error; - kfree(attr_name); - out_kfree_ops: - kfree(ops); - out_dput: - dput(dentry); - return -error; + if (copy_to_user(arg, &ageo, sizeof(ageo))) + return -EFAULT; + return 0; } -int -xfs_ioc_space( - struct xfs_inode *ip, - struct inode *inode, - struct file *filp, - int ioflags, - unsigned int cmd, - xfs_flock64_t *bf) +STATIC int +xfs_ioc_rtgroup_geometry( + struct xfs_mount *mp, + void __user *arg) { - int attr_flags = 0; + struct xfs_rtgroup *rtg; + struct xfs_rtgroup_geometry rgeo; int error; - /* - * Only allow the sys admin to reserve space unless - * unwritten extents are enabled. - */ - if (!xfs_sb_version_hasextflgbit(&ip->i_mount->m_sb) && - !capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); + if (copy_from_user(&rgeo, arg, sizeof(rgeo))) + return -EFAULT; + if (rgeo.rg_flags) + return -EINVAL; + if (memchr_inv(&rgeo.rg_reserved, 0, sizeof(rgeo.rg_reserved))) + return -EINVAL; + if (!xfs_has_rtgroups(mp)) + return -EINVAL; + + rtg = xfs_rtgroup_get(mp, rgeo.rg_number); + if (!rtg) + return -EINVAL; + + error = xfs_rtgroup_get_geometry(rtg, &rgeo); + xfs_rtgroup_put(rtg); + if (error) + return error; - if (inode->i_flags & (S_IMMUTABLE|S_APPEND)) - return -XFS_ERROR(EPERM); + if (copy_to_user(arg, &rgeo, sizeof(rgeo))) + return -EFAULT; + return 0; +} - if (!(filp->f_mode & FMODE_WRITE)) - return -XFS_ERROR(EBADF); +/* + * Linux extended inode flags interface. + */ - if (!S_ISREG(inode->i_mode)) - return -XFS_ERROR(EINVAL); +static void +xfs_fill_fsxattr( + struct xfs_inode *ip, + int whichfork, + struct file_kattr *fa) +{ + struct xfs_mount *mp = ip->i_mount; + struct xfs_ifork *ifp = xfs_ifork_ptr(ip, whichfork); - if (filp->f_flags & (O_NDELAY|O_NONBLOCK)) - attr_flags |= XFS_ATTR_NONBLOCK; + fileattr_fill_xflags(fa, xfs_ip2xflags(ip)); - if (filp->f_flags & O_DSYNC) - attr_flags |= XFS_ATTR_SYNC; + if (ip->i_diflags & XFS_DIFLAG_EXTSIZE) { + fa->fsx_extsize = XFS_FSB_TO_B(mp, ip->i_extsize); + } else if (ip->i_diflags & XFS_DIFLAG_EXTSZINHERIT) { + /* + * Don't let a misaligned extent size hint on a directory + * escape to userspace if it won't pass the setattr checks + * later. + */ + if ((ip->i_diflags & XFS_DIFLAG_RTINHERIT) && + xfs_extlen_to_rtxmod(mp, ip->i_extsize) > 0) { + fa->fsx_xflags &= ~(FS_XFLAG_EXTSIZE | + FS_XFLAG_EXTSZINHERIT); + fa->fsx_extsize = 0; + } else { + fa->fsx_extsize = XFS_FSB_TO_B(mp, ip->i_extsize); + } + } - if (ioflags & IO_INVIS) - attr_flags |= XFS_ATTR_DMI; + if (ip->i_diflags2 & XFS_DIFLAG2_COWEXTSIZE) { + /* + * Don't let a misaligned CoW extent size hint on a directory + * escape to userspace if it won't pass the setattr checks + * later. + */ + if ((ip->i_diflags & XFS_DIFLAG_RTINHERIT) && + ip->i_cowextsize % mp->m_sb.sb_rextsize > 0) { + fa->fsx_xflags &= ~FS_XFLAG_COWEXTSIZE; + fa->fsx_cowextsize = 0; + } else { + fa->fsx_cowextsize = XFS_FSB_TO_B(mp, ip->i_cowextsize); + } + } - error = mnt_want_write_file(filp); - if (error) - return error; - error = xfs_change_file_space(ip, cmd, bf, filp->f_pos, attr_flags); - mnt_drop_write_file(filp); - return -error; + fa->fsx_projid = ip->i_projid; + if (ifp && !xfs_need_iread_extents(ifp)) + fa->fsx_nextents = xfs_iext_count(ifp); + else + fa->fsx_nextents = xfs_ifork_nextents(ifp); } STATIC int -xfs_ioc_bulkstat( - xfs_mount_t *mp, - unsigned int cmd, +xfs_ioc_fsgetxattra( + xfs_inode_t *ip, void __user *arg) { - xfs_fsop_bulkreq_t bulkreq; - int count; /* # of records returned */ - xfs_ino_t inlast; /* last inode number */ - int done; - int error; - - /* done = 1 if there are more stats to get and if bulkstat */ - /* should be called again (unused here, but used in dmapi) */ + struct file_kattr fa; - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; + xfs_ilock(ip, XFS_ILOCK_SHARED); + xfs_fill_fsxattr(ip, XFS_ATTR_FORK, &fa); + xfs_iunlock(ip, XFS_ILOCK_SHARED); - if (XFS_FORCED_SHUTDOWN(mp)) - return -XFS_ERROR(EIO); + return copy_fsxattr_to_user(&fa, arg); +} - if (copy_from_user(&bulkreq, arg, sizeof(xfs_fsop_bulkreq_t))) - return -XFS_ERROR(EFAULT); +int +xfs_fileattr_get( + struct dentry *dentry, + struct file_kattr *fa) +{ + struct xfs_inode *ip = XFS_I(d_inode(dentry)); - if (copy_from_user(&inlast, bulkreq.lastip, sizeof(__s64))) - return -XFS_ERROR(EFAULT); + xfs_ilock(ip, XFS_ILOCK_SHARED); + xfs_fill_fsxattr(ip, XFS_DATA_FORK, fa); + xfs_iunlock(ip, XFS_ILOCK_SHARED); - if ((count = bulkreq.icount) <= 0) - return -XFS_ERROR(EINVAL); + return 0; +} - if (bulkreq.ubuffer == NULL) - return -XFS_ERROR(EINVAL); - - if (cmd == XFS_IOC_FSINUMBERS) - error = xfs_inumbers(mp, &inlast, &count, - bulkreq.ubuffer, xfs_inumbers_fmt); - else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE) - error = xfs_bulkstat_single(mp, &inlast, - bulkreq.ubuffer, &done); - else /* XFS_IOC_FSBULKSTAT */ - error = xfs_bulkstat(mp, &inlast, &count, xfs_bulkstat_one, - sizeof(xfs_bstat_t), bulkreq.ubuffer, - &done); +static int +xfs_ioctl_setattr_xflags( + struct xfs_trans *tp, + struct xfs_inode *ip, + struct file_kattr *fa) +{ + struct xfs_mount *mp = ip->i_mount; + bool rtflag = (fa->fsx_xflags & FS_XFLAG_REALTIME); + uint64_t i_flags2; - if (error) - return -error; + if (rtflag != XFS_IS_REALTIME_INODE(ip)) { + /* Can't change realtime flag if any extents are allocated. */ + if (xfs_inode_has_filedata(ip)) + return -EINVAL; - if (bulkreq.ocount != NULL) { - if (copy_to_user(bulkreq.lastip, &inlast, - sizeof(xfs_ino_t))) - return -XFS_ERROR(EFAULT); + /* + * If S_DAX is enabled on this file, we can only switch the + * device if both support fsdax. We can't update S_DAX because + * there might be other threads walking down the access paths. + */ + if (IS_DAX(VFS_I(ip)) && + (mp->m_ddev_targp->bt_daxdev == NULL || + (mp->m_rtdev_targp && + mp->m_rtdev_targp->bt_daxdev == NULL))) + return -EINVAL; + } - if (copy_to_user(bulkreq.ocount, &count, sizeof(count))) - return -XFS_ERROR(EFAULT); + if (rtflag) { + /* If realtime flag is set then must have realtime device */ + if (mp->m_sb.sb_rblocks == 0 || mp->m_sb.sb_rextsize == 0 || + xfs_extlen_to_rtxmod(mp, ip->i_extsize)) + return -EINVAL; } - return 0; -} + /* diflags2 only valid for v3 inodes. */ + i_flags2 = xfs_flags2diflags2(ip, fa->fsx_xflags); + if (i_flags2 && !xfs_has_v3inodes(mp)) + return -EINVAL; -STATIC int -xfs_ioc_fsgeometry_v1( - xfs_mount_t *mp, - void __user *arg) -{ - xfs_fsop_geom_t fsgeo; - int error; + ip->i_diflags = xfs_flags2diflags(ip, fa->fsx_xflags); + ip->i_diflags2 = i_flags2; - error = xfs_fs_geometry(mp, &fsgeo, 3); - if (error) - return -error; + xfs_diflags_to_iflags(ip, false); /* - * Caller should have passed an argument of type - * xfs_fsop_geom_v1_t. This is a proper subset of the - * xfs_fsop_geom_t that xfs_fs_geometry() fills in. + * Make the stable writes flag match that of the device the inode + * resides on when flipping the RT flag. */ - if (copy_to_user(arg, &fsgeo, sizeof(xfs_fsop_geom_v1_t))) - return -XFS_ERROR(EFAULT); + if (rtflag != XFS_IS_REALTIME_INODE(ip) && S_ISREG(VFS_I(ip)->i_mode)) + xfs_update_stable_writes(ip); + + xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG); + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + XFS_STATS_INC(mp, xs_ig_attrchg); return 0; } -STATIC int -xfs_ioc_fsgeometry( - xfs_mount_t *mp, - void __user *arg) +static void +xfs_ioctl_setattr_prepare_dax( + struct xfs_inode *ip, + struct file_kattr *fa) { - xfs_fsop_geom_t fsgeo; - int error; + struct xfs_mount *mp = ip->i_mount; + struct inode *inode = VFS_I(ip); - error = xfs_fs_geometry(mp, &fsgeo, 4); - if (error) - return -error; + if (S_ISDIR(inode->i_mode)) + return; - if (copy_to_user(arg, &fsgeo, sizeof(fsgeo))) - return -XFS_ERROR(EFAULT); - return 0; + if (xfs_has_dax_always(mp) || xfs_has_dax_never(mp)) + return; + + if (((fa->fsx_xflags & FS_XFLAG_DAX) && + !(ip->i_diflags2 & XFS_DIFLAG2_DAX)) || + (!(fa->fsx_xflags & FS_XFLAG_DAX) && + (ip->i_diflags2 & XFS_DIFLAG2_DAX))) + d_mark_dontcache(inode); } /* - * Linux extended inode flags interface. + * Set up the transaction structure for the setattr operation, checking that we + * have permission to do so. On success, return a clean transaction and the + * inode locked exclusively ready for further operation specific checks. On + * failure, return an error without modifying or locking the inode. */ - -STATIC unsigned int -xfs_merge_ioc_xflags( - unsigned int flags, - unsigned int start) +static struct xfs_trans * +xfs_ioctl_setattr_get_trans( + struct xfs_inode *ip, + struct xfs_dquot *pdqp) { - unsigned int xflags = start; + struct xfs_mount *mp = ip->i_mount; + struct xfs_trans *tp; + int error = -EROFS; - if (flags & FS_IMMUTABLE_FL) - xflags |= XFS_XFLAG_IMMUTABLE; - else - xflags &= ~XFS_XFLAG_IMMUTABLE; - if (flags & FS_APPEND_FL) - xflags |= XFS_XFLAG_APPEND; - else - xflags &= ~XFS_XFLAG_APPEND; - if (flags & FS_SYNC_FL) - xflags |= XFS_XFLAG_SYNC; - else - xflags &= ~XFS_XFLAG_SYNC; - if (flags & FS_NOATIME_FL) - xflags |= XFS_XFLAG_NOATIME; - else - xflags &= ~XFS_XFLAG_NOATIME; - if (flags & FS_NODUMP_FL) - xflags |= XFS_XFLAG_NODUMP; - else - xflags &= ~XFS_XFLAG_NODUMP; + if (xfs_is_readonly(mp)) + goto out_error; + error = -EIO; + if (xfs_is_shutdown(mp)) + goto out_error; - return xflags; -} + error = xfs_trans_alloc_ichange(ip, NULL, NULL, pdqp, + has_capability_noaudit(current, CAP_FOWNER), &tp); + if (error) + goto out_error; -STATIC unsigned int -xfs_di2lxflags( - __uint16_t di_flags) -{ - unsigned int flags = 0; - - if (di_flags & XFS_DIFLAG_IMMUTABLE) - flags |= FS_IMMUTABLE_FL; - if (di_flags & XFS_DIFLAG_APPEND) - flags |= FS_APPEND_FL; - if (di_flags & XFS_DIFLAG_SYNC) - flags |= FS_SYNC_FL; - if (di_flags & XFS_DIFLAG_NOATIME) - flags |= FS_NOATIME_FL; - if (di_flags & XFS_DIFLAG_NODUMP) - flags |= FS_NODUMP_FL; - return flags; + if (xfs_has_wsync(mp)) + xfs_trans_set_sync(tp); + + return tp; + +out_error: + return ERR_PTR(error); } -STATIC int -xfs_ioc_fsgetxattr( - xfs_inode_t *ip, - int attr, - void __user *arg) +/* + * Validate a proposed extent size hint. For regular files, the hint can only + * be changed if no extents are allocated. + */ +static int +xfs_ioctl_setattr_check_extsize( + struct xfs_inode *ip, + struct file_kattr *fa) { - struct fsxattr fa; + struct xfs_mount *mp = ip->i_mount; + xfs_failaddr_t failaddr; + uint16_t new_diflags; - memset(&fa, 0, sizeof(struct fsxattr)); + if (!fa->fsx_valid) + return 0; - xfs_ilock(ip, XFS_ILOCK_SHARED); - fa.fsx_xflags = xfs_ip2xflags(ip); - fa.fsx_extsize = ip->i_d.di_extsize << ip->i_mount->m_sb.sb_blocklog; - fa.fsx_projid = xfs_get_projid(ip); - - if (attr) { - if (ip->i_afp) { - if (ip->i_afp->if_flags & XFS_IFEXTENTS) - fa.fsx_nextents = ip->i_afp->if_bytes / - sizeof(xfs_bmbt_rec_t); - else - fa.fsx_nextents = ip->i_d.di_anextents; - } else - fa.fsx_nextents = 0; - } else { - if (ip->i_df.if_flags & XFS_IFEXTENTS) - fa.fsx_nextents = ip->i_df.if_bytes / - sizeof(xfs_bmbt_rec_t); - else - fa.fsx_nextents = ip->i_d.di_nextents; + if (S_ISREG(VFS_I(ip)->i_mode) && xfs_inode_has_filedata(ip) && + XFS_FSB_TO_B(mp, ip->i_extsize) != fa->fsx_extsize) + return -EINVAL; + + if (fa->fsx_extsize & mp->m_blockmask) + return -EINVAL; + + new_diflags = xfs_flags2diflags(ip, fa->fsx_xflags); + + /* + * Inode verifiers do not check that the extent size hint is an integer + * multiple of the rt extent size on a directory with both rtinherit + * and extszinherit flags set. Don't let sysadmins misconfigure + * directories. + */ + if ((new_diflags & XFS_DIFLAG_RTINHERIT) && + (new_diflags & XFS_DIFLAG_EXTSZINHERIT)) { + unsigned int rtextsize_bytes; + + rtextsize_bytes = XFS_FSB_TO_B(mp, mp->m_sb.sb_rextsize); + if (fa->fsx_extsize % rtextsize_bytes) + return -EINVAL; } - xfs_iunlock(ip, XFS_ILOCK_SHARED); - if (copy_to_user(arg, &fa, sizeof(fa))) - return -EFAULT; - return 0; + failaddr = xfs_inode_validate_extsize(ip->i_mount, + XFS_B_TO_FSB(mp, fa->fsx_extsize), + VFS_I(ip)->i_mode, new_diflags); + return failaddr != NULL ? -EINVAL : 0; } -STATIC void -xfs_set_diflags( +static int +xfs_ioctl_setattr_check_cowextsize( struct xfs_inode *ip, - unsigned int xflags) + struct file_kattr *fa) { - unsigned int di_flags; - - /* can't set PREALLOC this way, just preserve it */ - di_flags = (ip->i_d.di_flags & XFS_DIFLAG_PREALLOC); - if (xflags & XFS_XFLAG_IMMUTABLE) - di_flags |= XFS_DIFLAG_IMMUTABLE; - if (xflags & XFS_XFLAG_APPEND) - di_flags |= XFS_DIFLAG_APPEND; - if (xflags & XFS_XFLAG_SYNC) - di_flags |= XFS_DIFLAG_SYNC; - if (xflags & XFS_XFLAG_NOATIME) - di_flags |= XFS_DIFLAG_NOATIME; - if (xflags & XFS_XFLAG_NODUMP) - di_flags |= XFS_DIFLAG_NODUMP; - if (xflags & XFS_XFLAG_PROJINHERIT) - di_flags |= XFS_DIFLAG_PROJINHERIT; - if (xflags & XFS_XFLAG_NODEFRAG) - di_flags |= XFS_DIFLAG_NODEFRAG; - if (xflags & XFS_XFLAG_FILESTREAM) - di_flags |= XFS_DIFLAG_FILESTREAM; - if (S_ISDIR(ip->i_d.di_mode)) { - if (xflags & XFS_XFLAG_RTINHERIT) - di_flags |= XFS_DIFLAG_RTINHERIT; - if (xflags & XFS_XFLAG_NOSYMLINKS) - di_flags |= XFS_DIFLAG_NOSYMLINKS; - if (xflags & XFS_XFLAG_EXTSZINHERIT) - di_flags |= XFS_DIFLAG_EXTSZINHERIT; - } else if (S_ISREG(ip->i_d.di_mode)) { - if (xflags & XFS_XFLAG_REALTIME) - di_flags |= XFS_DIFLAG_REALTIME; - if (xflags & XFS_XFLAG_EXTSIZE) - di_flags |= XFS_DIFLAG_EXTSIZE; - } + struct xfs_mount *mp = ip->i_mount; + xfs_failaddr_t failaddr; + uint64_t new_diflags2; + uint16_t new_diflags; - ip->i_d.di_flags = di_flags; + if (!fa->fsx_valid) + return 0; + + if (fa->fsx_cowextsize & mp->m_blockmask) + return -EINVAL; + + new_diflags = xfs_flags2diflags(ip, fa->fsx_xflags); + new_diflags2 = xfs_flags2diflags2(ip, fa->fsx_xflags); + + failaddr = xfs_inode_validate_cowextsize(ip->i_mount, + XFS_B_TO_FSB(mp, fa->fsx_cowextsize), + VFS_I(ip)->i_mode, new_diflags, new_diflags2); + return failaddr != NULL ? -EINVAL : 0; } -STATIC void -xfs_diflags_to_linux( - struct xfs_inode *ip) +static int +xfs_ioctl_setattr_check_projid( + struct xfs_inode *ip, + struct file_kattr *fa) { - struct inode *inode = VFS_I(ip); - unsigned int xflags = xfs_ip2xflags(ip); + if (!fa->fsx_valid) + return 0; - if (xflags & XFS_XFLAG_IMMUTABLE) - inode->i_flags |= S_IMMUTABLE; - else - inode->i_flags &= ~S_IMMUTABLE; - if (xflags & XFS_XFLAG_APPEND) - inode->i_flags |= S_APPEND; - else - inode->i_flags &= ~S_APPEND; - if (xflags & XFS_XFLAG_SYNC) - inode->i_flags |= S_SYNC; - else - inode->i_flags &= ~S_SYNC; - if (xflags & XFS_XFLAG_NOATIME) - inode->i_flags |= S_NOATIME; - else - inode->i_flags &= ~S_NOATIME; + /* Disallow 32bit project ids if 32bit IDs are not enabled. */ + if (fa->fsx_projid > (uint16_t)-1 && + !xfs_has_projid32(ip->i_mount)) + return -EINVAL; + return 0; } -#define FSX_PROJID 1 -#define FSX_EXTSIZE 2 -#define FSX_XFLAGS 4 -#define FSX_NONBLOCK 8 - -STATIC int -xfs_ioctl_setattr( - xfs_inode_t *ip, - struct fsxattr *fa, - int mask) +int +xfs_fileattr_set( + struct mnt_idmap *idmap, + struct dentry *dentry, + struct file_kattr *fa) { + struct xfs_inode *ip = XFS_I(d_inode(dentry)); struct xfs_mount *mp = ip->i_mount; struct xfs_trans *tp; - unsigned int lock_flags = 0; - struct xfs_dquot *udqp = NULL; struct xfs_dquot *pdqp = NULL; struct xfs_dquot *olddquot = NULL; - int code; + int error; trace_xfs_ioctl_setattr(ip); - if (mp->m_flags & XFS_MOUNT_RDONLY) - return XFS_ERROR(EROFS); - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); + if (!fa->fsx_valid) { + if (fa->flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | + FS_NOATIME_FL | FS_NODUMP_FL | + FS_SYNC_FL | FS_DAX_FL | FS_PROJINHERIT_FL)) + return -EOPNOTSUPP; + } - /* - * Disallow 32bit project ids when projid32bit feature is not enabled. - */ - if ((mask & FSX_PROJID) && (fa->fsx_projid > (__uint16_t)-1) && - !xfs_sb_version_hasprojid32bit(&ip->i_mount->m_sb)) - return XFS_ERROR(EINVAL); + error = xfs_ioctl_setattr_check_projid(ip, fa); + if (error) + return error; /* * If disk quotas is on, we make sure that the dquots do exist on disk, @@ -954,381 +752,422 @@ xfs_ioctl_setattr( * If the IDs do change before we take the ilock, we're covered * because the i_*dquot fields will get updated anyway. */ - if (XFS_IS_QUOTA_ON(mp) && (mask & FSX_PROJID)) { - code = xfs_qm_vop_dqalloc(ip, ip->i_d.di_uid, - ip->i_d.di_gid, fa->fsx_projid, - XFS_QMOPT_PQUOTA, &udqp, NULL, &pdqp); - if (code) - return code; + if (fa->fsx_valid && XFS_IS_QUOTA_ON(mp)) { + error = xfs_qm_vop_dqalloc(ip, VFS_I(ip)->i_uid, + VFS_I(ip)->i_gid, fa->fsx_projid, + XFS_QMOPT_PQUOTA, NULL, NULL, &pdqp); + if (error) + return error; } - /* - * For the other attributes, we acquire the inode lock and - * first do an error checking pass. - */ - tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE); - code = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0); - if (code) - goto error_return; - - lock_flags = XFS_ILOCK_EXCL; - xfs_ilock(ip, lock_flags); + xfs_ioctl_setattr_prepare_dax(ip, fa); - /* - * CAP_FOWNER overrides the following restrictions: - * - * The user ID of the calling process must be equal - * to the file owner ID, except in cases where the - * CAP_FSETID capability is applicable. - */ - if (current_fsuid() != ip->i_d.di_uid && !capable(CAP_FOWNER)) { - code = XFS_ERROR(EPERM); - goto error_return; + tp = xfs_ioctl_setattr_get_trans(ip, pdqp); + if (IS_ERR(tp)) { + error = PTR_ERR(tp); + goto error_free_dquots; } - /* - * Do a quota reservation only if projid is actually going to change. - */ - if (mask & FSX_PROJID) { - if (XFS_IS_QUOTA_RUNNING(mp) && - XFS_IS_PQUOTA_ON(mp) && - xfs_get_projid(ip) != fa->fsx_projid) { - ASSERT(tp); - code = xfs_qm_vop_chown_reserve(tp, ip, udqp, NULL, - pdqp, capable(CAP_FOWNER) ? - XFS_QMOPT_FORCE_RES : 0); - if (code) /* out of quota */ - goto error_return; - } - } - - if (mask & FSX_EXTSIZE) { - /* - * Can't change extent size if any extents are allocated. - */ - if (ip->i_d.di_nextents && - ((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) != - fa->fsx_extsize)) { - code = XFS_ERROR(EINVAL); /* EFBIG? */ - goto error_return; - } + error = xfs_ioctl_setattr_check_extsize(ip, fa); + if (error) + goto error_trans_cancel; - /* - * Extent size must be a multiple of the appropriate block - * size, if set at all. It must also be smaller than the - * maximum extent size supported by the filesystem. - * - * Also, for non-realtime files, limit the extent size hint to - * half the size of the AGs in the filesystem so alignment - * doesn't result in extents larger than an AG. - */ - if (fa->fsx_extsize != 0) { - xfs_extlen_t size; - xfs_fsblock_t extsize_fsb; - - extsize_fsb = XFS_B_TO_FSB(mp, fa->fsx_extsize); - if (extsize_fsb > MAXEXTLEN) { - code = XFS_ERROR(EINVAL); - goto error_return; - } - - if (XFS_IS_REALTIME_INODE(ip) || - ((mask & FSX_XFLAGS) && - (fa->fsx_xflags & XFS_XFLAG_REALTIME))) { - size = mp->m_sb.sb_rextsize << - mp->m_sb.sb_blocklog; - } else { - size = mp->m_sb.sb_blocksize; - if (extsize_fsb > mp->m_sb.sb_agblocks / 2) { - code = XFS_ERROR(EINVAL); - goto error_return; - } - } - - if (fa->fsx_extsize % size) { - code = XFS_ERROR(EINVAL); - goto error_return; - } - } - } + error = xfs_ioctl_setattr_check_cowextsize(ip, fa); + if (error) + goto error_trans_cancel; + error = xfs_ioctl_setattr_xflags(tp, ip, fa); + if (error) + goto error_trans_cancel; - if (mask & FSX_XFLAGS) { - /* - * Can't change realtime flag if any extents are allocated. - */ - if ((ip->i_d.di_nextents || ip->i_delayed_blks) && - (XFS_IS_REALTIME_INODE(ip)) != - (fa->fsx_xflags & XFS_XFLAG_REALTIME)) { - code = XFS_ERROR(EINVAL); /* EFBIG? */ - goto error_return; - } + if (!fa->fsx_valid) + goto skip_xattr; + /* + * Change file ownership. Must be the owner or privileged. CAP_FSETID + * overrides the following restrictions: + * + * The set-user-ID and set-group-ID bits of a file will be cleared upon + * successful return from chown() + */ - /* - * If realtime flag is set then must have realtime data. - */ - if ((fa->fsx_xflags & XFS_XFLAG_REALTIME)) { - if ((mp->m_sb.sb_rblocks == 0) || - (mp->m_sb.sb_rextsize == 0) || - (ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) { - code = XFS_ERROR(EINVAL); - goto error_return; - } - } + if ((VFS_I(ip)->i_mode & (S_ISUID|S_ISGID)) && + !capable_wrt_inode_uidgid(idmap, VFS_I(ip), CAP_FSETID)) + VFS_I(ip)->i_mode &= ~(S_ISUID|S_ISGID); - /* - * Can't modify an immutable/append-only file unless - * we have appropriate permission. - */ - if ((ip->i_d.di_flags & - (XFS_DIFLAG_IMMUTABLE|XFS_DIFLAG_APPEND) || - (fa->fsx_xflags & - (XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) && - !capable(CAP_LINUX_IMMUTABLE)) { - code = XFS_ERROR(EPERM); - goto error_return; + /* Change the ownerships and register project quota modifications */ + if (ip->i_projid != fa->fsx_projid) { + if (XFS_IS_PQUOTA_ON(mp)) { + olddquot = xfs_qm_vop_chown(tp, ip, + &ip->i_pdquot, pdqp); } + ip->i_projid = fa->fsx_projid; } - xfs_trans_ijoin(tp, ip, 0); - /* - * Change file ownership. Must be the owner or privileged. + * Only set the extent size hint if we've already determined that the + * extent size hint should be set on the inode. If no extent size flags + * are set on the inode then unconditionally clear the extent size hint. */ - if (mask & FSX_PROJID) { - /* - * CAP_FSETID overrides the following restrictions: - * - * The set-user-ID and set-group-ID bits of a file will be - * cleared upon successful return from chown() - */ - if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) && - !capable(CAP_FSETID)) - ip->i_d.di_mode &= ~(S_ISUID|S_ISGID); - - /* - * Change the ownerships and register quota modifications - * in the transaction. - */ - if (xfs_get_projid(ip) != fa->fsx_projid) { - if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp)) { - olddquot = xfs_qm_vop_chown(tp, ip, - &ip->i_pdquot, pdqp); - } - xfs_set_projid(ip, fa->fsx_projid); - - /* - * We may have to rev the inode as well as - * the superblock version number since projids didn't - * exist before DINODE_VERSION_2 and SB_VERSION_NLINK. - */ - if (ip->i_d.di_version == 1) - xfs_bump_ino_vers2(tp, ip); - } - - } + if (ip->i_diflags & (XFS_DIFLAG_EXTSIZE | XFS_DIFLAG_EXTSZINHERIT)) + ip->i_extsize = XFS_B_TO_FSB(mp, fa->fsx_extsize); + else + ip->i_extsize = 0; - if (mask & FSX_EXTSIZE) - ip->i_d.di_extsize = fa->fsx_extsize >> mp->m_sb.sb_blocklog; - if (mask & FSX_XFLAGS) { - xfs_set_diflags(ip, fa->fsx_xflags); - xfs_diflags_to_linux(ip); + if (xfs_has_v3inodes(mp)) { + if (ip->i_diflags2 & XFS_DIFLAG2_COWEXTSIZE) + ip->i_cowextsize = XFS_B_TO_FSB(mp, fa->fsx_cowextsize); + else + ip->i_cowextsize = 0; } - xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG); - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - - XFS_STATS_INC(xs_ig_attrchg); - - /* - * If this is a synchronous mount, make sure that the - * transaction goes to disk before returning to the user. - * This is slightly sub-optimal in that truncates require - * two sync transactions instead of one for wsync filesystems. - * One for the truncate and one for the timestamps since we - * don't want to change the timestamps unless we're sure the - * truncate worked. Truncates are less than 1% of the laddis - * mix so this probably isn't worth the trouble to optimize. - */ - if (mp->m_flags & XFS_MOUNT_WSYNC) - xfs_trans_set_sync(tp); - code = xfs_trans_commit(tp, 0); - xfs_iunlock(ip, lock_flags); +skip_xattr: + error = xfs_trans_commit(tp); /* * Release any dquot(s) the inode had kept before chown. */ xfs_qm_dqrele(olddquot); - xfs_qm_dqrele(udqp); xfs_qm_dqrele(pdqp); - return code; + return error; - error_return: - xfs_qm_dqrele(udqp); +error_trans_cancel: + xfs_trans_cancel(tp); +error_free_dquots: xfs_qm_dqrele(pdqp); - xfs_trans_cancel(tp, 0); - if (lock_flags) - xfs_iunlock(ip, lock_flags); - return code; + return error; +} + +static bool +xfs_getbmap_format( + struct kgetbmap *p, + struct getbmapx __user *u, + size_t recsize) +{ + if (put_user(p->bmv_offset, &u->bmv_offset) || + put_user(p->bmv_block, &u->bmv_block) || + put_user(p->bmv_length, &u->bmv_length) || + put_user(0, &u->bmv_count) || + put_user(0, &u->bmv_entries)) + return false; + if (recsize < sizeof(struct getbmapx)) + return true; + if (put_user(0, &u->bmv_iflags) || + put_user(p->bmv_oflags, &u->bmv_oflags) || + put_user(0, &u->bmv_unused1) || + put_user(0, &u->bmv_unused2)) + return false; + return true; } STATIC int -xfs_ioc_fssetxattr( - xfs_inode_t *ip, - struct file *filp, +xfs_ioc_getbmap( + struct file *file, + unsigned int cmd, void __user *arg) { - struct fsxattr fa; - unsigned int mask; - int error; + struct getbmapx bmx = { 0 }; + struct kgetbmap *buf; + size_t recsize; + int error, i; + + switch (cmd) { + case XFS_IOC_GETBMAPA: + bmx.bmv_iflags = BMV_IF_ATTRFORK; + fallthrough; + case XFS_IOC_GETBMAP: + /* struct getbmap is a strict subset of struct getbmapx. */ + recsize = sizeof(struct getbmap); + break; + case XFS_IOC_GETBMAPX: + recsize = sizeof(struct getbmapx); + break; + default: + return -EINVAL; + } - if (copy_from_user(&fa, arg, sizeof(fa))) + if (copy_from_user(&bmx, arg, recsize)) return -EFAULT; - mask = FSX_XFLAGS | FSX_EXTSIZE | FSX_PROJID; - if (filp->f_flags & (O_NDELAY|O_NONBLOCK)) - mask |= FSX_NONBLOCK; + if (bmx.bmv_count < 2) + return -EINVAL; + if (bmx.bmv_count >= INT_MAX / recsize) + return -ENOMEM; - error = mnt_want_write_file(filp); + buf = kvcalloc(bmx.bmv_count, sizeof(*buf), GFP_KERNEL); + if (!buf) + return -ENOMEM; + + error = xfs_getbmap(XFS_I(file_inode(file)), &bmx, buf); if (error) - return error; - error = xfs_ioctl_setattr(ip, &fa, mask); - mnt_drop_write_file(filp); - return -error; -} + goto out_free_buf; -STATIC int -xfs_ioc_getxflags( - xfs_inode_t *ip, - void __user *arg) -{ - unsigned int flags; + error = -EFAULT; + if (copy_to_user(arg, &bmx, recsize)) + goto out_free_buf; + arg += recsize; + + for (i = 0; i < bmx.bmv_entries; i++) { + if (!xfs_getbmap_format(buf + i, arg, recsize)) + goto out_free_buf; + arg += recsize; + } - flags = xfs_di2lxflags(ip->i_d.di_flags); - if (copy_to_user(arg, &flags, sizeof(flags))) - return -EFAULT; - return 0; + error = 0; +out_free_buf: + kvfree(buf); + return error; } -STATIC int -xfs_ioc_setxflags( - xfs_inode_t *ip, - struct file *filp, - void __user *arg) +int +xfs_ioc_swapext( + xfs_swapext_t *sxp) { - struct fsxattr fa; - unsigned int flags; - unsigned int mask; - int error; + xfs_inode_t *ip, *tip; - if (copy_from_user(&flags, arg, sizeof(flags))) - return -EFAULT; + /* Pull information for the target fd */ + CLASS(fd, f)((int)sxp->sx_fdtarget); + if (fd_empty(f)) + return -EINVAL; - if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \ - FS_NOATIME_FL | FS_NODUMP_FL | \ - FS_SYNC_FL)) - return -EOPNOTSUPP; + if (!(fd_file(f)->f_mode & FMODE_WRITE) || + !(fd_file(f)->f_mode & FMODE_READ) || + (fd_file(f)->f_flags & O_APPEND)) + return -EBADF; - mask = FSX_XFLAGS; - if (filp->f_flags & (O_NDELAY|O_NONBLOCK)) - mask |= FSX_NONBLOCK; - fa.fsx_xflags = xfs_merge_ioc_xflags(flags, xfs_ip2xflags(ip)); + CLASS(fd, tmp)((int)sxp->sx_fdtmp); + if (fd_empty(tmp)) + return -EINVAL; - error = mnt_want_write_file(filp); - if (error) - return error; - error = xfs_ioctl_setattr(ip, &fa, mask); - mnt_drop_write_file(filp); - return -error; + if (!(fd_file(tmp)->f_mode & FMODE_WRITE) || + !(fd_file(tmp)->f_mode & FMODE_READ) || + (fd_file(tmp)->f_flags & O_APPEND)) + return -EBADF; + + if (IS_SWAPFILE(file_inode(fd_file(f))) || + IS_SWAPFILE(file_inode(fd_file(tmp)))) + return -EINVAL; + + /* + * We need to ensure that the fds passed in point to XFS inodes + * before we cast and access them as XFS structures as we have no + * control over what the user passes us here. + */ + if (fd_file(f)->f_op != &xfs_file_operations || + fd_file(tmp)->f_op != &xfs_file_operations) + return -EINVAL; + + ip = XFS_I(file_inode(fd_file(f))); + tip = XFS_I(file_inode(fd_file(tmp))); + + if (ip->i_mount != tip->i_mount) + return -EINVAL; + + if (ip->i_ino == tip->i_ino) + return -EINVAL; + + if (xfs_is_shutdown(ip->i_mount)) + return -EIO; + + return xfs_swap_extents(ip, tip, sxp); } -STATIC int -xfs_getbmap_format(void **ap, struct getbmapx *bmv, int *full) +static int +xfs_ioc_getlabel( + struct xfs_mount *mp, + char __user *user_label) { - struct getbmap __user *base = *ap; + struct xfs_sb *sbp = &mp->m_sb; + char label[XFSLABEL_MAX + 1]; + + /* Paranoia */ + BUILD_BUG_ON(sizeof(sbp->sb_fname) > FSLABEL_MAX); - /* copy only getbmap portion (not getbmapx) */ - if (copy_to_user(base, bmv, sizeof(struct getbmap))) - return XFS_ERROR(EFAULT); + /* 1 larger than sb_fname, so this ensures a trailing NUL char */ + spin_lock(&mp->m_sb_lock); + memtostr_pad(label, sbp->sb_fname); + spin_unlock(&mp->m_sb_lock); - *ap += sizeof(struct getbmap); + if (copy_to_user(user_label, label, sizeof(label))) + return -EFAULT; return 0; } -STATIC int -xfs_ioc_getbmap( - struct xfs_inode *ip, - int ioflags, - unsigned int cmd, - void __user *arg) +static int +xfs_ioc_setlabel( + struct file *filp, + struct xfs_mount *mp, + char __user *newlabel) { - struct getbmapx bmx; + struct xfs_sb *sbp = &mp->m_sb; + char label[XFSLABEL_MAX + 1]; + size_t len; int error; - if (copy_from_user(&bmx, arg, sizeof(struct getbmapx))) - return -XFS_ERROR(EFAULT); + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + /* + * The generic ioctl allows up to FSLABEL_MAX chars, but XFS is much + * smaller, at 12 bytes. We copy one more to be sure we find the + * (required) NULL character to test the incoming label length. + * NB: The on disk label doesn't need to be null terminated. + */ + if (copy_from_user(label, newlabel, XFSLABEL_MAX + 1)) + return -EFAULT; + len = strnlen(label, XFSLABEL_MAX + 1); + if (len > sizeof(sbp->sb_fname)) + return -EINVAL; - if (bmx.bmv_count < 2) - return -XFS_ERROR(EINVAL); + error = mnt_want_write_file(filp); + if (error) + return error; - bmx.bmv_iflags = (cmd == XFS_IOC_GETBMAPA ? BMV_IF_ATTRFORK : 0); - if (ioflags & IO_INVIS) - bmx.bmv_iflags |= BMV_IF_NO_DMAPI_READ; + spin_lock(&mp->m_sb_lock); + memset(sbp->sb_fname, 0, sizeof(sbp->sb_fname)); + memcpy(sbp->sb_fname, label, len); + spin_unlock(&mp->m_sb_lock); - error = xfs_getbmap(ip, &bmx, xfs_getbmap_format, - (struct getbmap *)arg+1); + /* + * Now we do several things to satisfy userspace. + * In addition to normal logging of the primary superblock, we also + * immediately write these changes to sector zero for the primary, then + * update all backup supers (as xfs_db does for a label change), then + * invalidate the block device page cache. This is so that any prior + * buffered reads from userspace (i.e. from blkid) are invalidated, + * and userspace will see the newly-written label. + */ + error = xfs_sync_sb_buf(mp, true); if (error) - return -error; + goto out; + /* + * growfs also updates backup supers so lock against that. + */ + mutex_lock(&mp->m_growlock); + error = xfs_update_secondary_sbs(mp); + mutex_unlock(&mp->m_growlock); - /* copy back header - only size of getbmap */ - if (copy_to_user(arg, &bmx, sizeof(struct getbmap))) - return -XFS_ERROR(EFAULT); - return 0; + invalidate_bdev(mp->m_ddev_targp->bt_bdev); + if (xfs_has_rtsb(mp) && mp->m_rtdev_targp) + invalidate_bdev(mp->m_rtdev_targp->bt_bdev); + +out: + mnt_drop_write_file(filp); + return error; } -STATIC int -xfs_getbmapx_format(void **ap, struct getbmapx *bmv, int *full) +static inline int +xfs_fs_eofblocks_from_user( + struct xfs_fs_eofblocks *src, + struct xfs_icwalk *dst) { - struct getbmapx __user *base = *ap; - - if (copy_to_user(base, bmv, sizeof(struct getbmapx))) - return XFS_ERROR(EFAULT); + if (src->eof_version != XFS_EOFBLOCKS_VERSION) + return -EINVAL; + + if (src->eof_flags & ~XFS_EOF_FLAGS_VALID) + return -EINVAL; + + if (memchr_inv(&src->pad32, 0, sizeof(src->pad32)) || + memchr_inv(src->pad64, 0, sizeof(src->pad64))) + return -EINVAL; + + dst->icw_flags = 0; + if (src->eof_flags & XFS_EOF_FLAGS_SYNC) + dst->icw_flags |= XFS_ICWALK_FLAG_SYNC; + if (src->eof_flags & XFS_EOF_FLAGS_UID) + dst->icw_flags |= XFS_ICWALK_FLAG_UID; + if (src->eof_flags & XFS_EOF_FLAGS_GID) + dst->icw_flags |= XFS_ICWALK_FLAG_GID; + if (src->eof_flags & XFS_EOF_FLAGS_PRID) + dst->icw_flags |= XFS_ICWALK_FLAG_PRID; + if (src->eof_flags & XFS_EOF_FLAGS_MINFILESIZE) + dst->icw_flags |= XFS_ICWALK_FLAG_MINFILESIZE; + + dst->icw_prid = src->eof_prid; + dst->icw_min_file_size = src->eof_min_file_size; + + dst->icw_uid = INVALID_UID; + if (src->eof_flags & XFS_EOF_FLAGS_UID) { + dst->icw_uid = make_kuid(current_user_ns(), src->eof_uid); + if (!uid_valid(dst->icw_uid)) + return -EINVAL; + } - *ap += sizeof(struct getbmapx); + dst->icw_gid = INVALID_GID; + if (src->eof_flags & XFS_EOF_FLAGS_GID) { + dst->icw_gid = make_kgid(current_user_ns(), src->eof_gid); + if (!gid_valid(dst->icw_gid)) + return -EINVAL; + } return 0; } -STATIC int -xfs_ioc_getbmapx( - struct xfs_inode *ip, - void __user *arg) +static int +xfs_ioctl_getset_resblocks( + struct file *filp, + unsigned int cmd, + void __user *arg) { - struct getbmapx bmx; + struct xfs_mount *mp = XFS_I(file_inode(filp))->i_mount; + struct xfs_fsop_resblks fsop = { }; int error; - if (copy_from_user(&bmx, arg, sizeof(bmx))) - return -XFS_ERROR(EFAULT); + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; - if (bmx.bmv_count < 2) - return -XFS_ERROR(EINVAL); + if (cmd == XFS_IOC_SET_RESBLKS) { + if (xfs_is_readonly(mp)) + return -EROFS; - if (bmx.bmv_iflags & (~BMV_IF_VALID)) - return -XFS_ERROR(EINVAL); + if (copy_from_user(&fsop, arg, sizeof(fsop))) + return -EFAULT; - error = xfs_getbmap(ip, &bmx, xfs_getbmapx_format, - (struct getbmapx *)arg+1); - if (error) - return -error; + error = mnt_want_write_file(filp); + if (error) + return error; + error = xfs_reserve_blocks(mp, XC_FREE_BLOCKS, fsop.resblks); + mnt_drop_write_file(filp); + if (error) + return error; + } + + spin_lock(&mp->m_sb_lock); + fsop.resblks = mp->m_free[XC_FREE_BLOCKS].res_total; + fsop.resblks_avail = mp->m_free[XC_FREE_BLOCKS].res_avail; + spin_unlock(&mp->m_sb_lock); - /* copy back header */ - if (copy_to_user(arg, &bmx, sizeof(struct getbmapx))) - return -XFS_ERROR(EFAULT); + if (copy_to_user(arg, &fsop, sizeof(fsop))) + return -EFAULT; + return 0; +} +static int +xfs_ioctl_fs_counts( + struct xfs_mount *mp, + struct xfs_fsop_counts __user *uarg) +{ + struct xfs_fsop_counts out = { + .allocino = percpu_counter_read_positive(&mp->m_icount), + .freeino = percpu_counter_read_positive(&mp->m_ifree), + .freedata = xfs_estimate_freecounter(mp, XC_FREE_BLOCKS) - + xfs_freecounter_unavailable(mp, XC_FREE_BLOCKS), + .freertx = xfs_estimate_freecounter(mp, XC_FREE_RTEXTENTS), + }; + + if (copy_to_user(uarg, &out, sizeof(out))) + return -EFAULT; return 0; } /* + * These long-unused ioctls were removed from the official ioctl API in 5.17, + * but retain these definitions so that we can log warnings about them. + */ +#define XFS_IOC_ALLOCSP _IOW ('X', 10, struct xfs_flock64) +#define XFS_IOC_FREESP _IOW ('X', 11, struct xfs_flock64) +#define XFS_IOC_ALLOCSP64 _IOW ('X', 36, struct xfs_flock64) +#define XFS_IOC_FREESP64 _IOW ('X', 37, struct xfs_flock64) + +/* * Note: some of the ioctl's return positive numbers as a * byte count indicating success, such as readlink_by_handle. * So we don't "sign flip" like most other routines. This means @@ -1344,93 +1183,89 @@ xfs_file_ioctl( struct xfs_inode *ip = XFS_I(inode); struct xfs_mount *mp = ip->i_mount; void __user *arg = (void __user *)p; - int ioflags = 0; int error; - if (filp->f_mode & FMODE_NOCMTIME) - ioflags |= IO_INVIS; - trace_xfs_file_ioctl(ip); switch (cmd) { case FITRIM: return xfs_ioc_trim(mp, arg); + case FS_IOC_GETFSLABEL: + return xfs_ioc_getlabel(mp, arg); + case FS_IOC_SETFSLABEL: + return xfs_ioc_setlabel(filp, mp, arg); case XFS_IOC_ALLOCSP: case XFS_IOC_FREESP: - case XFS_IOC_RESVSP: - case XFS_IOC_UNRESVSP: case XFS_IOC_ALLOCSP64: case XFS_IOC_FREESP64: - case XFS_IOC_RESVSP64: - case XFS_IOC_UNRESVSP64: - case XFS_IOC_ZERO_RANGE: { - xfs_flock64_t bf; - - if (copy_from_user(&bf, arg, sizeof(bf))) - return -XFS_ERROR(EFAULT); - return xfs_ioc_space(ip, inode, filp, ioflags, cmd, &bf); - } + xfs_warn_once(mp, + "%s should use fallocate; XFS_IOC_{ALLOC,FREE}SP ioctl unsupported", + current->comm); + return -ENOTTY; case XFS_IOC_DIOINFO: { - struct dioattr da; - xfs_buftarg_t *target = - XFS_IS_REALTIME_INODE(ip) ? - mp->m_rtdev_targp : mp->m_ddev_targp; + struct kstat st; + struct dioattr da; - da.d_mem = da.d_miniosz = 1 << target->bt_sshift; - da.d_maxiosz = INT_MAX & ~(da.d_miniosz - 1); + error = vfs_getattr(&filp->f_path, &st, STATX_DIOALIGN, 0); + if (error) + return error; + /* + * Some userspace directly feeds the return value to + * posix_memalign, which fails for values that are smaller than + * the pointer size. Round up the value to not break userspace. + */ + da.d_mem = roundup(st.dio_mem_align, sizeof(void *)); + da.d_miniosz = st.dio_offset_align; + da.d_maxiosz = INT_MAX & ~(da.d_miniosz - 1); if (copy_to_user(arg, &da, sizeof(da))) - return -XFS_ERROR(EFAULT); + return -EFAULT; return 0; } case XFS_IOC_FSBULKSTAT_SINGLE: case XFS_IOC_FSBULKSTAT: case XFS_IOC_FSINUMBERS: - return xfs_ioc_bulkstat(mp, cmd, arg); + return xfs_ioc_fsbulkstat(filp, cmd, arg); - case XFS_IOC_FSGEOMETRY_V1: - return xfs_ioc_fsgeometry_v1(mp, arg); + case XFS_IOC_BULKSTAT: + return xfs_ioc_bulkstat(filp, cmd, arg); + case XFS_IOC_INUMBERS: + return xfs_ioc_inumbers(mp, cmd, arg); + case XFS_IOC_FSGEOMETRY_V1: + return xfs_ioc_fsgeometry(mp, arg, 3); + case XFS_IOC_FSGEOMETRY_V4: + return xfs_ioc_fsgeometry(mp, arg, 4); case XFS_IOC_FSGEOMETRY: - return xfs_ioc_fsgeometry(mp, arg); + return xfs_ioc_fsgeometry(mp, arg, 5); + + case XFS_IOC_AG_GEOMETRY: + return xfs_ioc_ag_geometry(mp, arg); + case XFS_IOC_RTGROUP_GEOMETRY: + return xfs_ioc_rtgroup_geometry(mp, arg); case XFS_IOC_GETVERSION: return put_user(inode->i_generation, (int __user *)arg); - case XFS_IOC_FSGETXATTR: - return xfs_ioc_fsgetxattr(ip, 0, arg); case XFS_IOC_FSGETXATTRA: - return xfs_ioc_fsgetxattr(ip, 1, arg); - case XFS_IOC_FSSETXATTR: - return xfs_ioc_fssetxattr(ip, filp, arg); - case XFS_IOC_GETXFLAGS: - return xfs_ioc_getxflags(ip, arg); - case XFS_IOC_SETXFLAGS: - return xfs_ioc_setxflags(ip, filp, arg); - - case XFS_IOC_FSSETDM: { - struct fsdmidata dmi; - - if (copy_from_user(&dmi, arg, sizeof(dmi))) - return -XFS_ERROR(EFAULT); - - error = mnt_want_write_file(filp); - if (error) - return error; - - error = xfs_set_dmattrs(ip, dmi.fsd_dmevmask, - dmi.fsd_dmstate); - mnt_drop_write_file(filp); - return -error; - } - + return xfs_ioc_fsgetxattra(ip, arg); + case XFS_IOC_GETPARENTS: + return xfs_ioc_getparents(filp, arg); + case XFS_IOC_GETPARENTS_BY_HANDLE: + return xfs_ioc_getparents_by_handle(filp, arg); case XFS_IOC_GETBMAP: case XFS_IOC_GETBMAPA: - return xfs_ioc_getbmap(ip, ioflags, cmd, arg); - case XFS_IOC_GETBMAPX: - return xfs_ioc_getbmapx(ip, arg); + return xfs_ioc_getbmap(filp, cmd, arg); + + case FS_IOC_GETFSMAP: + return xfs_ioc_getfsmap(ip, arg); + + case XFS_IOC_SCRUBV_METADATA: + return xfs_ioc_scrubv_metadata(filp, arg); + case XFS_IOC_SCRUB_METADATA: + return xfs_ioc_scrub_metadata(filp, arg); case XFS_IOC_FD_TO_HANDLE: case XFS_IOC_PATH_TO_HANDLE: @@ -1438,24 +1273,22 @@ xfs_file_ioctl( xfs_fsop_handlereq_t hreq; if (copy_from_user(&hreq, arg, sizeof(hreq))) - return -XFS_ERROR(EFAULT); + return -EFAULT; return xfs_find_handle(cmd, &hreq); } case XFS_IOC_OPEN_BY_HANDLE: { xfs_fsop_handlereq_t hreq; if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t))) - return -XFS_ERROR(EFAULT); + return -EFAULT; return xfs_open_by_handle(filp, &hreq); } - case XFS_IOC_FSSETDM_BY_HANDLE: - return xfs_fssetdm_by_handle(filp, arg); case XFS_IOC_READLINK_BY_HANDLE: { xfs_fsop_handlereq_t hreq; if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t))) - return -XFS_ERROR(EFAULT); + return -EFAULT; return xfs_readlink_by_handle(filp, &hreq); } case XFS_IOC_ATTRLIST_BY_HANDLE: @@ -1468,125 +1301,74 @@ xfs_file_ioctl( struct xfs_swapext sxp; if (copy_from_user(&sxp, arg, sizeof(xfs_swapext_t))) - return -XFS_ERROR(EFAULT); + return -EFAULT; error = mnt_want_write_file(filp); if (error) return error; - error = xfs_swapext(&sxp); + error = xfs_ioc_swapext(&sxp); mnt_drop_write_file(filp); - return -error; - } - - case XFS_IOC_FSCOUNTS: { - xfs_fsop_counts_t out; - - error = xfs_fs_counts(mp, &out); - if (error) - return -error; - - if (copy_to_user(arg, &out, sizeof(out))) - return -XFS_ERROR(EFAULT); - return 0; - } - - case XFS_IOC_SET_RESBLKS: { - xfs_fsop_resblks_t inout; - __uint64_t in; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - if (mp->m_flags & XFS_MOUNT_RDONLY) - return -XFS_ERROR(EROFS); - - if (copy_from_user(&inout, arg, sizeof(inout))) - return -XFS_ERROR(EFAULT); - - error = mnt_want_write_file(filp); - if (error) - return error; - - /* input parameter is passed in resblks field of structure */ - in = inout.resblks; - error = xfs_reserve_blocks(mp, &in, &inout); - mnt_drop_write_file(filp); - if (error) - return -error; - - if (copy_to_user(arg, &inout, sizeof(inout))) - return -XFS_ERROR(EFAULT); - return 0; + return error; } - case XFS_IOC_GET_RESBLKS: { - xfs_fsop_resblks_t out; + case XFS_IOC_FSCOUNTS: + return xfs_ioctl_fs_counts(mp, arg); - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - error = xfs_reserve_blocks(mp, NULL, &out); - if (error) - return -error; - - if (copy_to_user(arg, &out, sizeof(out))) - return -XFS_ERROR(EFAULT); - - return 0; - } + case XFS_IOC_SET_RESBLKS: + case XFS_IOC_GET_RESBLKS: + return xfs_ioctl_getset_resblocks(filp, cmd, arg); case XFS_IOC_FSGROWFSDATA: { - xfs_growfs_data_t in; + struct xfs_growfs_data in; if (copy_from_user(&in, arg, sizeof(in))) - return -XFS_ERROR(EFAULT); + return -EFAULT; error = mnt_want_write_file(filp); if (error) return error; error = xfs_growfs_data(mp, &in); mnt_drop_write_file(filp); - return -error; + return error; } case XFS_IOC_FSGROWFSLOG: { - xfs_growfs_log_t in; + struct xfs_growfs_log in; if (copy_from_user(&in, arg, sizeof(in))) - return -XFS_ERROR(EFAULT); + return -EFAULT; error = mnt_want_write_file(filp); if (error) return error; error = xfs_growfs_log(mp, &in); mnt_drop_write_file(filp); - return -error; + return error; } case XFS_IOC_FSGROWFSRT: { xfs_growfs_rt_t in; if (copy_from_user(&in, arg, sizeof(in))) - return -XFS_ERROR(EFAULT); + return -EFAULT; error = mnt_want_write_file(filp); if (error) return error; error = xfs_growfs_rt(mp, &in); mnt_drop_write_file(filp); - return -error; + return error; } case XFS_IOC_GOINGDOWN: { - __uint32_t in; + uint32_t in; if (!capable(CAP_SYS_ADMIN)) return -EPERM; - if (get_user(in, (__uint32_t __user *)arg)) - return -XFS_ERROR(EFAULT); + if (get_user(in, (uint32_t __user *)arg)) + return -EFAULT; - error = xfs_fs_goingdown(mp, in); - return -error; + return xfs_fs_goingdown(mp, in); } case XFS_IOC_ERROR_INJECTION: { @@ -1596,39 +1378,47 @@ xfs_file_ioctl( return -EPERM; if (copy_from_user(&in, arg, sizeof(in))) - return -XFS_ERROR(EFAULT); + return -EFAULT; - error = xfs_errortag_add(in.errtag, mp); - return -error; + return xfs_errortag_add(mp, in.errtag); } case XFS_IOC_ERROR_CLEARALL: if (!capable(CAP_SYS_ADMIN)) return -EPERM; - error = xfs_errortag_clearall(mp, 1); - return -error; + return xfs_errortag_clearall(mp); case XFS_IOC_FREE_EOFBLOCKS: { - struct xfs_eofblocks eofb; + struct xfs_fs_eofblocks eofb; + struct xfs_icwalk icw; - if (copy_from_user(&eofb, arg, sizeof(eofb))) - return -XFS_ERROR(EFAULT); + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (xfs_is_readonly(mp)) + return -EROFS; - if (eofb.eof_version != XFS_EOFBLOCKS_VERSION) - return -XFS_ERROR(EINVAL); + if (copy_from_user(&eofb, arg, sizeof(eofb))) + return -EFAULT; - if (eofb.eof_flags & ~XFS_EOF_FLAGS_VALID) - return -XFS_ERROR(EINVAL); + error = xfs_fs_eofblocks_from_user(&eofb, &icw); + if (error) + return error; - if (memchr_inv(&eofb.pad32, 0, sizeof(eofb.pad32)) || - memchr_inv(eofb.pad64, 0, sizeof(eofb.pad64))) - return -XFS_ERROR(EINVAL); + trace_xfs_ioc_free_eofblocks(mp, &icw, _RET_IP_); - error = xfs_icache_free_eofblocks(mp, &eofb); - return -error; + guard(super_write)(mp->m_super); + return xfs_blockgc_free_space(mp, &icw); } + case XFS_IOC_EXCHANGE_RANGE: + return xfs_ioc_exchange_range(filp, arg); + case XFS_IOC_START_COMMIT: + return xfs_ioc_start_commit(filp, arg); + case XFS_IOC_COMMIT_RANGE: + return xfs_ioc_commit_range(filp, arg); + default: return -ENOTTY; } |
