#include "xfs_inode.h"
#include "xfs_error.h"
#include "xfs_rw.h"
-#include "xfs_ioctl32.h"
+#include "xfs_vnodeops.h"
+#include "xfs_da_btree.h"
+#include "xfs_ioctl.h"
#include <linux/dcache.h>
#include <linux/smp_lock.h>
static struct vm_operations_struct xfs_file_vm_ops;
-#ifdef CONFIG_XFS_DMAPI
-static struct vm_operations_struct xfs_dmapi_file_vm_ops;
-#endif
-
-STATIC inline ssize_t
-__xfs_file_read(
- struct kiocb *iocb,
- const struct iovec *iov,
- unsigned long nr_segs,
- int ioflags,
- loff_t pos)
-{
- struct file *file = iocb->ki_filp;
- bhv_vnode_t *vp = vn_from_inode(file->f_dentry->d_inode);
-
- BUG_ON(iocb->ki_pos != pos);
- if (unlikely(file->f_flags & O_DIRECT))
- ioflags |= IO_ISDIRECT;
- return bhv_vop_read(vp, iocb, iov, nr_segs, &iocb->ki_pos,
- ioflags, NULL);
-}
STATIC ssize_t
xfs_file_aio_read(
unsigned long nr_segs,
loff_t pos)
{
- return __xfs_file_read(iocb, iov, nr_segs, IO_ISAIO, pos);
-}
-
-STATIC ssize_t
-xfs_file_aio_read_invis(
- struct kiocb *iocb,
- const struct iovec *iov,
- unsigned long nr_segs,
- loff_t pos)
-{
- return __xfs_file_read(iocb, iov, nr_segs, IO_ISAIO|IO_INVIS, pos);
-}
-
-STATIC inline ssize_t
-__xfs_file_write(
- struct kiocb *iocb,
- const struct iovec *iov,
- unsigned long nr_segs,
- int ioflags,
- loff_t pos)
-{
- struct file *file = iocb->ki_filp;
- struct inode *inode = file->f_mapping->host;
- bhv_vnode_t *vp = vn_from_inode(inode);
+ struct file *file = iocb->ki_filp;
+ int ioflags = IO_ISAIO;
BUG_ON(iocb->ki_pos != pos);
if (unlikely(file->f_flags & O_DIRECT))
ioflags |= IO_ISDIRECT;
- return bhv_vop_write(vp, iocb, iov, nr_segs, &iocb->ki_pos,
- ioflags, NULL);
+ if (file->f_mode & FMODE_NOCMTIME)
+ ioflags |= IO_INVIS;
+ return xfs_read(XFS_I(file->f_path.dentry->d_inode), iocb, iov,
+ nr_segs, &iocb->ki_pos, ioflags);
}
STATIC ssize_t
unsigned long nr_segs,
loff_t pos)
{
- return __xfs_file_write(iocb, iov, nr_segs, IO_ISAIO, pos);
-}
-
-STATIC ssize_t
-xfs_file_aio_write_invis(
- struct kiocb *iocb,
- const struct iovec *iov,
- unsigned long nr_segs,
- loff_t pos)
-{
- return __xfs_file_write(iocb, iov, nr_segs, IO_ISAIO|IO_INVIS, pos);
-}
-
-STATIC inline ssize_t
-__xfs_file_readv(
- struct file *file,
- const struct iovec *iov,
- int ioflags,
- unsigned long nr_segs,
- loff_t *ppos)
-{
- struct inode *inode = file->f_mapping->host;
- bhv_vnode_t *vp = vn_from_inode(inode);
- struct kiocb kiocb;
- ssize_t rval;
-
- init_sync_kiocb(&kiocb, file);
- kiocb.ki_pos = *ppos;
-
- if (unlikely(file->f_flags & O_DIRECT))
- ioflags |= IO_ISDIRECT;
- rval = bhv_vop_read(vp, &kiocb, iov, nr_segs,
- &kiocb.ki_pos, ioflags, NULL);
-
- *ppos = kiocb.ki_pos;
- return rval;
-}
-
-STATIC ssize_t
-xfs_file_readv(
- struct file *file,
- const struct iovec *iov,
- unsigned long nr_segs,
- loff_t *ppos)
-{
- return __xfs_file_readv(file, iov, 0, nr_segs, ppos);
-}
-
-STATIC ssize_t
-xfs_file_readv_invis(
- struct file *file,
- const struct iovec *iov,
- unsigned long nr_segs,
- loff_t *ppos)
-{
- return __xfs_file_readv(file, iov, IO_INVIS, nr_segs, ppos);
-}
-
-STATIC inline ssize_t
-__xfs_file_writev(
- struct file *file,
- const struct iovec *iov,
- int ioflags,
- unsigned long nr_segs,
- loff_t *ppos)
-{
- struct inode *inode = file->f_mapping->host;
- bhv_vnode_t *vp = vn_from_inode(inode);
- struct kiocb kiocb;
- ssize_t rval;
+ struct file *file = iocb->ki_filp;
+ int ioflags = IO_ISAIO;
- init_sync_kiocb(&kiocb, file);
- kiocb.ki_pos = *ppos;
+ BUG_ON(iocb->ki_pos != pos);
if (unlikely(file->f_flags & O_DIRECT))
ioflags |= IO_ISDIRECT;
-
- rval = bhv_vop_write(vp, &kiocb, iov, nr_segs,
- &kiocb.ki_pos, ioflags, NULL);
-
- *ppos = kiocb.ki_pos;
- return rval;
-}
-
-STATIC ssize_t
-xfs_file_writev(
- struct file *file,
- const struct iovec *iov,
- unsigned long nr_segs,
- loff_t *ppos)
-{
- return __xfs_file_writev(file, iov, 0, nr_segs, ppos);
-}
-
-STATIC ssize_t
-xfs_file_writev_invis(
- struct file *file,
- const struct iovec *iov,
- unsigned long nr_segs,
- loff_t *ppos)
-{
- return __xfs_file_writev(file, iov, IO_INVIS, nr_segs, ppos);
-}
-
-STATIC ssize_t
-xfs_file_sendfile(
- struct file *filp,
- loff_t *pos,
- size_t count,
- read_actor_t actor,
- void *target)
-{
- return bhv_vop_sendfile(vn_from_inode(filp->f_dentry->d_inode),
- filp, pos, 0, count, actor, target, NULL);
-}
-
-STATIC ssize_t
-xfs_file_sendfile_invis(
- struct file *filp,
- loff_t *pos,
- size_t count,
- read_actor_t actor,
- void *target)
-{
- return bhv_vop_sendfile(vn_from_inode(filp->f_dentry->d_inode),
- filp, pos, IO_INVIS, count, actor, target, NULL);
+ if (file->f_mode & FMODE_NOCMTIME)
+ ioflags |= IO_INVIS;
+ return xfs_write(XFS_I(file->f_mapping->host), iocb, iov, nr_segs,
+ &iocb->ki_pos, ioflags);
}
STATIC ssize_t
size_t len,
unsigned int flags)
{
- return bhv_vop_splice_read(vn_from_inode(infilp->f_dentry->d_inode),
- infilp, ppos, pipe, len, flags, 0, NULL);
-}
+ int ioflags = 0;
-STATIC ssize_t
-xfs_file_splice_read_invis(
- struct file *infilp,
- loff_t *ppos,
- struct pipe_inode_info *pipe,
- size_t len,
- unsigned int flags)
-{
- return bhv_vop_splice_read(vn_from_inode(infilp->f_dentry->d_inode),
- infilp, ppos, pipe, len, flags, IO_INVIS,
- NULL);
+ if (infilp->f_mode & FMODE_NOCMTIME)
+ ioflags |= IO_INVIS;
+
+ return xfs_splice_read(XFS_I(infilp->f_path.dentry->d_inode),
+ infilp, ppos, pipe, len, flags, ioflags);
}
STATIC ssize_t
size_t len,
unsigned int flags)
{
- return bhv_vop_splice_write(vn_from_inode(outfilp->f_dentry->d_inode),
- pipe, outfilp, ppos, len, flags, 0, NULL);
-}
+ int ioflags = 0;
-STATIC ssize_t
-xfs_file_splice_write_invis(
- struct pipe_inode_info *pipe,
- struct file *outfilp,
- loff_t *ppos,
- size_t len,
- unsigned int flags)
-{
- return bhv_vop_splice_write(vn_from_inode(outfilp->f_dentry->d_inode),
- pipe, outfilp, ppos, len, flags, IO_INVIS,
- NULL);
+ if (outfilp->f_mode & FMODE_NOCMTIME)
+ ioflags |= IO_INVIS;
+
+ return xfs_splice_write(XFS_I(outfilp->f_path.dentry->d_inode),
+ pipe, outfilp, ppos, len, flags, ioflags);
}
STATIC int
xfs_file_open(
struct inode *inode,
- struct file *filp)
+ struct file *file)
{
- if (!(filp->f_flags & O_LARGEFILE) && i_size_read(inode) > MAX_NON_LFS)
+ if (!(file->f_flags & O_LARGEFILE) && i_size_read(inode) > MAX_NON_LFS)
return -EFBIG;
- return -bhv_vop_open(vn_from_inode(inode), NULL);
+ if (XFS_FORCED_SHUTDOWN(XFS_M(inode->i_sb)))
+ return -EIO;
+ return 0;
}
STATIC int
-xfs_file_close(
- struct file *filp,
- fl_owner_t id)
+xfs_dir_open(
+ struct inode *inode,
+ struct file *file)
{
- return -bhv_vop_close(vn_from_inode(filp->f_dentry->d_inode), 0,
- file_count(filp) > 1 ? L_FALSE : L_TRUE, NULL);
+ struct xfs_inode *ip = XFS_I(inode);
+ int mode;
+ int error;
+
+ error = xfs_file_open(inode, file);
+ if (error)
+ return error;
+
+ /*
+ * If there are any blocks, read-ahead block 0 as we're almost
+ * certain to have the next operation be a read there.
+ */
+ mode = xfs_ilock_map_shared(ip);
+ if (ip->i_d.di_nextents > 0)
+ xfs_da_reada_buf(NULL, ip, 0, XFS_DATA_FORK);
+ xfs_iunlock(ip, mode);
+ return 0;
}
STATIC int
struct inode *inode,
struct file *filp)
{
- bhv_vnode_t *vp = vn_from_inode(inode);
-
- if (vp)
- return -bhv_vop_release(vp);
- return 0;
+ return -xfs_release(XFS_I(inode));
}
+/*
+ * We ignore the datasync flag here because a datasync is effectively
+ * identical to an fsync. That is, datasync implies that we need to write
+ * only the metadata needed to be able to access the data that is written
+ * if we crash after the call completes. Hence if we are writing beyond
+ * EOF we have to log the inode size change as well, which makes it a
+ * full fsync. If we don't write beyond EOF, the inode core will be
+ * clean in memory and so we don't need to log the inode, just like
+ * fsync.
+ */
STATIC int
xfs_file_fsync(
struct file *filp,
struct dentry *dentry,
int datasync)
{
- bhv_vnode_t *vp = vn_from_inode(dentry->d_inode);
- int flags = FSYNC_WAIT;
-
- if (datasync)
- flags |= FSYNC_DATA;
- if (VN_TRUNC(vp))
- VUNTRUNCATE(vp);
- return -bhv_vop_fsync(vp, flags, NULL, (xfs_off_t)0, (xfs_off_t)-1);
+ xfs_iflags_clear(XFS_I(dentry->d_inode), XFS_ITRUNCATED);
+ return -xfs_fsync(XFS_I(dentry->d_inode));
}
-#ifdef CONFIG_XFS_DMAPI
-STATIC struct page *
-xfs_vm_nopage(
- struct vm_area_struct *area,
- unsigned long address,
- int *type)
-{
- struct inode *inode = area->vm_file->f_dentry->d_inode;
- bhv_vnode_t *vp = vn_from_inode(inode);
-
- ASSERT_ALWAYS(vp->v_vfsp->vfs_flag & VFS_DMI);
- if (XFS_SEND_MMAP(XFS_VFSTOM(vp->v_vfsp), area, 0))
- return NULL;
- return filemap_nopage(area, address, type);
-}
-#endif /* CONFIG_XFS_DMAPI */
-
STATIC int
xfs_file_readdir(
struct file *filp,
void *dirent,
filldir_t filldir)
{
- int error = 0;
- bhv_vnode_t *vp = vn_from_inode(filp->f_dentry->d_inode);
- uio_t uio;
- iovec_t iov;
- int eof = 0;
- caddr_t read_buf;
- int namelen, size = 0;
- size_t rlen = PAGE_CACHE_SIZE;
- xfs_off_t start_offset, curr_offset;
- xfs_dirent_t *dbp = NULL;
-
- /* Try fairly hard to get memory */
- do {
- if ((read_buf = kmalloc(rlen, GFP_KERNEL)))
- break;
- rlen >>= 1;
- } while (rlen >= 1024);
-
- if (read_buf == NULL)
- return -ENOMEM;
-
- uio.uio_iov = &iov;
- uio.uio_segflg = UIO_SYSSPACE;
- curr_offset = filp->f_pos;
- if (filp->f_pos != 0x7fffffff)
- uio.uio_offset = filp->f_pos;
- else
- uio.uio_offset = 0xffffffff;
-
- while (!eof) {
- uio.uio_resid = iov.iov_len = rlen;
- iov.iov_base = read_buf;
- uio.uio_iovcnt = 1;
-
- start_offset = uio.uio_offset;
-
- error = bhv_vop_readdir(vp, &uio, NULL, &eof);
- if ((uio.uio_offset == start_offset) || error) {
- size = 0;
- break;
- }
-
- size = rlen - uio.uio_resid;
- dbp = (xfs_dirent_t *)read_buf;
- while (size > 0) {
- namelen = strlen(dbp->d_name);
-
- if (filldir(dirent, dbp->d_name, namelen,
- (loff_t) curr_offset & 0x7fffffff,
- (ino_t) dbp->d_ino,
- DT_UNKNOWN)) {
- goto done;
- }
- size -= dbp->d_reclen;
- curr_offset = (loff_t)dbp->d_off /* & 0x7fffffff */;
- dbp = (xfs_dirent_t *)((char *)dbp + dbp->d_reclen);
- }
- }
-done:
- if (!error) {
- if (size == 0)
- filp->f_pos = uio.uio_offset & 0x7fffffff;
- else if (dbp)
- filp->f_pos = curr_offset;
- }
+ struct inode *inode = filp->f_path.dentry->d_inode;
+ xfs_inode_t *ip = XFS_I(inode);
+ int error;
+ size_t bufsize;
+
+ /*
+ * The Linux API doesn't pass down the total size of the buffer
+ * we read into down to the filesystem. With the filldir concept
+ * it's not needed for correct information, but the XFS dir2 leaf
+ * code wants an estimate of the buffer size to calculate it's
+ * readahead window and size the buffers used for mapping to
+ * physical blocks.
+ *
+ * Try to give it an estimate that's good enough, maybe at some
+ * point we can change the ->readdir prototype to include the
+ * buffer size.
+ */
+ bufsize = (size_t)min_t(loff_t, PAGE_SIZE, ip->i_d.di_size);
- kfree(read_buf);
- return -error;
+ error = xfs_readdir(ip, dirent, bufsize,
+ (xfs_off_t *)&filp->f_pos, filldir);
+ if (error)
+ return -error;
+ return 0;
}
STATIC int
struct vm_area_struct *vma)
{
vma->vm_ops = &xfs_file_vm_ops;
-
-#ifdef CONFIG_XFS_DMAPI
- if (vn_from_inode(filp->f_dentry->d_inode)->v_vfsp->vfs_flag & VFS_DMI)
- vma->vm_ops = &xfs_dmapi_file_vm_ops;
-#endif /* CONFIG_XFS_DMAPI */
+ vma->vm_flags |= VM_CAN_NONLINEAR;
file_accessed(filp);
return 0;
}
-STATIC long
-xfs_file_ioctl(
- struct file *filp,
- unsigned int cmd,
- unsigned long p)
-{
- int error;
- struct inode *inode = filp->f_dentry->d_inode;
- bhv_vnode_t *vp = vn_from_inode(inode);
-
- error = bhv_vop_ioctl(vp, inode, filp, 0, cmd, (void __user *)p);
- VMODIFY(vp);
-
- /* NOTE: some of the ioctl's return positive #'s as a
- * byte count indicating success, such as
- * readlink_by_handle. So we don't "sign flip"
- * like most other routines. This means true
- * errors need to be returned as a negative value.
- */
- return error;
-}
-
-STATIC long
-xfs_file_ioctl_invis(
- struct file *filp,
- unsigned int cmd,
- unsigned long p)
-{
- int error;
- struct inode *inode = filp->f_dentry->d_inode;
- bhv_vnode_t *vp = vn_from_inode(inode);
-
- error = bhv_vop_ioctl(vp, inode, filp, IO_INVIS, cmd, (void __user *)p);
- VMODIFY(vp);
-
- /* NOTE: some of the ioctl's return positive #'s as a
- * byte count indicating success, such as
- * readlink_by_handle. So we don't "sign flip"
- * like most other routines. This means true
- * errors need to be returned as a negative value.
- */
- return error;
-}
-
-#ifdef CONFIG_XFS_DMAPI
-#ifdef HAVE_VMOP_MPROTECT
-STATIC int
-xfs_vm_mprotect(
- struct vm_area_struct *vma,
- unsigned int newflags)
-{
- bhv_vnode_t *vp = vn_from_inode(vma->vm_file->f_dentry->d_inode);
- int error = 0;
-
- if (vp->v_vfsp->vfs_flag & VFS_DMI) {
- if ((vma->vm_flags & VM_MAYSHARE) &&
- (newflags & VM_WRITE) && !(vma->vm_flags & VM_WRITE)) {
- xfs_mount_t *mp = XFS_VFSTOM(vp->v_vfsp);
-
- error = XFS_SEND_MMAP(mp, vma, VM_WRITE);
- }
- }
- return error;
-}
-#endif /* HAVE_VMOP_MPROTECT */
-#endif /* CONFIG_XFS_DMAPI */
-
-#ifdef HAVE_FOP_OPEN_EXEC
-/* If the user is attempting to execute a file that is offline then
- * we have to trigger a DMAPI READ event before the file is marked as busy
- * otherwise the invisible I/O will not be able to write to the file to bring
- * it back online.
+/*
+ * mmap()d file has taken write protection fault and is being made
+ * writable. We can set the page state up correctly for a writable
+ * page, which means we can do correct delalloc accounting (ENOSPC
+ * checking!) and unwritten extent mapping.
*/
STATIC int
-xfs_file_open_exec(
- struct inode *inode)
+xfs_vm_page_mkwrite(
+ struct vm_area_struct *vma,
+ struct page *page)
{
- bhv_vnode_t *vp = vn_from_inode(inode);
-
- if (unlikely(vp->v_vfsp->vfs_flag & VFS_DMI)) {
- xfs_mount_t *mp = XFS_VFSTOM(vp->v_vfsp);
- xfs_inode_t *ip = xfs_vtoi(vp);
-
- if (!ip)
- return -EINVAL;
- if (DM_EVENT_ENABLED(vp->v_vfsp, ip, DM_EVENT_READ))
- return -XFS_SEND_DATA(mp, DM_EVENT_READ, vp,
- 0, 0, 0, NULL);
- }
- return 0;
+ return block_page_mkwrite(vma, page, xfs_get_blocks);
}
-#endif /* HAVE_FOP_OPEN_EXEC */
const struct file_operations xfs_file_operations = {
.llseek = generic_file_llseek,
.read = do_sync_read,
.write = do_sync_write,
- .readv = xfs_file_readv,
- .writev = xfs_file_writev,
.aio_read = xfs_file_aio_read,
.aio_write = xfs_file_aio_write,
- .sendfile = xfs_file_sendfile,
.splice_read = xfs_file_splice_read,
.splice_write = xfs_file_splice_write,
.unlocked_ioctl = xfs_file_ioctl,
#endif
.mmap = xfs_file_mmap,
.open = xfs_file_open,
- .flush = xfs_file_close,
.release = xfs_file_release,
.fsync = xfs_file_fsync,
#ifdef HAVE_FOP_OPEN_EXEC
#endif
};
-const struct file_operations xfs_invis_file_operations = {
- .llseek = generic_file_llseek,
- .read = do_sync_read,
- .write = do_sync_write,
- .readv = xfs_file_readv_invis,
- .writev = xfs_file_writev_invis,
- .aio_read = xfs_file_aio_read_invis,
- .aio_write = xfs_file_aio_write_invis,
- .sendfile = xfs_file_sendfile_invis,
- .splice_read = xfs_file_splice_read_invis,
- .splice_write = xfs_file_splice_write_invis,
- .unlocked_ioctl = xfs_file_ioctl_invis,
-#ifdef CONFIG_COMPAT
- .compat_ioctl = xfs_file_compat_invis_ioctl,
-#endif
- .mmap = xfs_file_mmap,
- .open = xfs_file_open,
- .flush = xfs_file_close,
- .release = xfs_file_release,
- .fsync = xfs_file_fsync,
-};
-
-
const struct file_operations xfs_dir_file_operations = {
+ .open = xfs_dir_open,
.read = generic_read_dir,
.readdir = xfs_file_readdir,
+ .llseek = generic_file_llseek,
.unlocked_ioctl = xfs_file_ioctl,
#ifdef CONFIG_COMPAT
.compat_ioctl = xfs_file_compat_ioctl,
};
static struct vm_operations_struct xfs_file_vm_ops = {
- .nopage = filemap_nopage,
- .populate = filemap_populate,
-};
-
-#ifdef CONFIG_XFS_DMAPI
-static struct vm_operations_struct xfs_dmapi_file_vm_ops = {
- .nopage = xfs_vm_nopage,
- .populate = filemap_populate,
-#ifdef HAVE_VMOP_MPROTECT
- .mprotect = xfs_vm_mprotect,
-#endif
+ .fault = filemap_fault,
+ .page_mkwrite = xfs_vm_page_mkwrite,
};
-#endif /* CONFIG_XFS_DMAPI */