/*
* It's worth taking a moment to describe how mmap is implemented
* for zfs because it differs considerably from other Linux filesystems.
* However, this issue is handled the same way under OpenSolaris.
*
* The issue is that by design zfs bypasses the Linux page cache and
* leaves all caching up to the ARC. This has been shown to work
* well for the common read(2)/write(2) case. However, mmap(2)
* is problem because it relies on being tightly integrated with the
* page cache. To handle this we cache mmap'ed files twice, once in
* the ARC and a second time in the page cache. The code is careful
* to keep both copies synchronized.
*
* When a file with an mmap'ed region is written to using write(2)
* both the data in the ARC and existing pages in the page cache
* are updated. For a read(2) data will be read first from the page
* cache then the ARC if needed. Neither a write(2) or read(2) will
* will ever result in new pages being added to the page cache.
*
* New pages are added to the page cache only via .readpage() which
* is called when the vfs needs to read a page off disk to back the
* virtual memory region. These pages may be modified without
* notifying the ARC and will be written out periodically via
* .writepage(). This will occur due to either a sync or the usual
* page aging behavior. Note because a read(2) of a mmap'ed file
* will always check the page cache first even when the ARC is out
* of date correct data will still be returned.
*
* While this implementation ensures correct behavior it does have
* have some drawbacks. The most obvious of which is that it
* increases the required memory footprint when access mmap'ed
* files. It also adds additional complexity to the code keeping
* both caches synchronized.
*
* Longer term it may be possible to cleanly resolve this wart by
* mapping page cache pages directly on to the ARC buffers. The
* Linux address space operations are flexible enough to allow
* selection of which pages back a particular index. The trick
* would be working out the details of which subsystem is in
* charge, the ARC, the page cache, or both. It may also prove
* helpful to move the ARC buffers to a scatter-gather lists
* rather than a vmalloc'ed region.
*/
static int
zpl_mmap(struct file *filp, struct vm_area_struct *vma)
{
struct inode *ip = filp->f_mapping->host;
znode_t *zp = ITOZ(ip);
int error;
fstrans_cookie_t cookie;
cookie = spl_fstrans_mark();
error = -zfs_map(ip, vma->vm_pgoff, (caddr_t *)vma->vm_start,
(size_t)(vma->vm_end - vma->vm_start), vma->vm_flags);
spl_fstrans_unmark(cookie);
if (error)
return (error);
error = generic_file_mmap(filp, vma);
if (error)
return (error);
mutex_enter(&zp->z_lock);
zp->z_is_mapped = 1;
mutex_exit(&zp->z_lock);
return (error);
}
static int ext2_file_mmap(struct file *file, struct vm_area_struct *vma)
{
if (!IS_DAX(file_inode(file)))
return generic_file_mmap(file, vma);
file_accessed(file);
vma->vm_ops = &ext2_dax_vm_ops;
return 0;
}
static int fuse_file_mmap(struct file *file, struct vm_area_struct *vma)
{
if ((vma->vm_flags & VM_SHARED)) {
if ((vma->vm_flags & VM_WRITE))
return -ENODEV;
else
vma->vm_flags &= ~VM_MAYWRITE;
}
return generic_file_mmap(file, vma);
}
static int ext2_file_mmap(struct file *file, struct vm_area_struct *vma)
{
if (!IS_DAX(file_inode(file)))
return generic_file_mmap(file, vma);
file_accessed(file);
vma->vm_ops = &ext2_dax_vm_ops;
vma->vm_flags |= VM_MIXEDMAP | VM_HUGEPAGE;
return 0;
}
static int ecryptfs_file_mmap(struct file *file, struct vm_area_struct *vma)
{
int rc;
rc = generic_file_mmap(file, vma);
if (!rc)
vma->vm_ops = &ecryptfs_file_vm_ops;
return rc;
}
static int cramfs_mmap(struct file *file, struct vm_area_struct *vma)
{
unsigned long address, length;
struct inode *inode = file->f_dentry->d_inode;
struct super_block *sb = inode->i_sb;
/* this is only used in the case of read-only maps for XIP */
if (vma->vm_flags & VM_WRITE)
return generic_file_mmap(file, vma);
if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE))
return -EINVAL;
address = PAGE_ALIGN(sb->CRAMFS_SB_LINEAR_PHYS_ADDR + OFFSET(inode));
address += vma->vm_pgoff << PAGE_SHIFT;
length = vma->vm_end - vma->vm_start;
if (length > inode->i_size)
length = inode->i_size;
length = PAGE_ALIGN(length);
#if 0
/* Doing the following makes it slower and more broken. bdl */
/*
* Accessing memory above the top the kernel knows about or
* through a file pointer that was marked O_SYNC will be
* done non-cached.
*/
vma->vm_page_prot =
__pgprot((pgprot_val(vma->vm_page_prot) & ~_CACHE_MASK)
| _CACHE_UNCACHED);
#endif
/*
* Don't dump addresses that are not real memory to a core file.
*/
vma->vm_flags |= VM_IO;
flush_tlb_page(vma, address);
if (remap_page_range(vma->vm_start, address, length,
vma->vm_page_prot))
return -EAGAIN;
#ifdef DEBUG_CRAMFS_XIP
printk("cramfs_mmap: mapped %s at 0x%08lx, length %lu to vma 0x%08lx"
", page_prot 0x%08lx\n",
file->f_dentry->d_name.name, address, length,
vma->vm_start, pgprot_val(vma->vm_page_prot));
#endif
return 0;
}
int
xixfs_file_mmap(
struct file * file,
struct vm_area_struct * vma
)
{
DebugTrace(DEBUG_LEVEL_ERROR, (DEBUG_TARGET_FCB|DEBUG_TARGET_VFSAPIT),
("ENTER xixfs_file_mmap (%s).\n", file->f_dentry->d_name.name));
return generic_file_mmap(file, vma);
}
static int ecryptfs_mmap(struct file *file, struct vm_area_struct *vma)
{
struct file *lower_file = ecryptfs_file_to_lower(file);
/*
* Don't allow mmap on top of file systems that don't support it
* natively. If FILESYSTEM_MAX_STACK_DEPTH > 2 or ecryptfs
* allows recursive mounting, this will need to be extended.
*/
if (!lower_file->f_op->mmap)
return -ENODEV;
return generic_file_mmap(file, vma);
}
static int
v9fs_file_mmap(struct file *filp, struct vm_area_struct *vma)
{
int retval;
retval = generic_file_mmap(filp, vma);
if (!retval)
vma->vm_ops = &v9fs_file_vm_ops;
return retval;
}
static int
nfs_file_mmap(struct file * file, struct vm_area_struct * vma)
{
struct dentry *dentry = file->f_dentry;
int status;
dfprintk(VFS, "nfs: mmap(%s/%s)\n",
dentry->d_parent->d_name.name, dentry->d_name.name);
status = nfs_revalidate_inode(NFS_DSERVER(dentry), dentry);
if (!status)
status = generic_file_mmap(file, vma);
return status;
}
static int coda_file_mmap(struct file * file, struct vm_area_struct * vma)
{
struct coda_inode_info *cii;
int res;
coda_vfs_stat.file_mmap++;
ENTRY;
cii = ITOC(file->f_dentry->d_inode);
res =generic_file_mmap(file, vma);
EXIT;
return res;
}
static int
nfs_file_mmap(struct file * file, struct vm_area_struct * vma)
{
struct dentry *dentry = file->f_path.dentry;
struct inode *inode = dentry->d_inode;
int status;
dfprintk(VFS, "nfs: mmap(%s/%s)\n",
dentry->d_parent->d_name.name, dentry->d_name.name);
status = nfs_revalidate_mapping(inode, file->f_mapping);
if (!status)
status = generic_file_mmap(file, vma);
return status;
}
int
nfs_file_mmap(struct file * file, struct vm_area_struct * vma)
{
struct inode *inode = file_inode(file);
int status;
dprintk("NFS: mmap(%pD2)\n", file);
/* Note: generic_file_mmap() returns ENOSYS on nommu systems
* so we call that before revalidating the mapping
*/
status = generic_file_mmap(file, vma);
if (!status) {
vma->vm_ops = &nfs_file_vm_ops;
status = nfs_revalidate_mapping(inode, file->f_mapping);
}
return status;
}
static int
smb_file_mmap(struct file * file, struct vm_area_struct * vma)
{
struct dentry * dentry = file->f_dentry;
int status;
VERBOSE("file %s/%s, address %lu - %lu\n",
DENTRY_PATH(dentry), vma->vm_start, vma->vm_end);
status = smb_revalidate_inode(dentry);
if (status) {
PARANOIA("%s/%s validation failed, error=%d\n",
DENTRY_PATH(dentry), status);
goto out;
}
status = generic_file_mmap(file, vma);
out:
return status;
}
int ll_file_mmap(struct file *file, struct vm_area_struct *vma)
{
struct inode *inode = file_inode(file);
int rc;
if (ll_file_nolock(file))
return -EOPNOTSUPP;
ll_stats_ops_tally(ll_i2sbi(inode), LPROC_LL_MAP, 1);
rc = generic_file_mmap(file, vma);
if (rc == 0) {
vma->vm_ops = &ll_file_vm_ops;
vma->vm_ops->open(vma);
/* update the inode's size and mtime */
rc = ll_glimpse_size(inode);
}
return rc;
}
static int
nfs_file_mmap(struct file * file, struct vm_area_struct * vma)
{
struct dentry *dentry = file->f_path.dentry;
struct inode *inode = dentry->d_inode;
int status;
dprintk("NFS: mmap(%s/%s)\n",
dentry->d_parent->d_name.name, dentry->d_name.name);
/* Note: generic_file_mmap() returns ENOSYS on nommu systems
* so we call that before revalidating the mapping
*/
status = generic_file_mmap(file, vma);
if (!status) {
vma->vm_ops = &nfs_file_vm_ops;
status = nfs_revalidate_mapping(inode, file->f_mapping);
}
return status;
}
int linvfs_generic_file_mmap(struct file *filp, struct vm_area_struct *vma)
{
vnode_t *vp;
int ret;
/* this will return a (-) error so flip */
ret = -generic_file_mmap(filp, vma);
if (!ret) {
vattr_t va, *vap;
vap = &va;
vap->va_mask = AT_UPDATIME;
vp = LINVFS_GET_VP(filp->f_dentry->d_inode);
ASSERT(vp);
VOP_SETATTR(vp, vap, AT_UPDATIME, NULL, ret);
}
return(-ret);
}
static int
HgfsMmap(struct file *file, // IN: File we operate on
struct vm_area_struct *vma) // IN/OUT: VM area information
{
int result;
ASSERT(file);
ASSERT(vma);
ASSERT(file->f_dentry);
LOG(6, (KERN_DEBUG "VMware hgfs: HgfsMmap: was called\n"));
result = HgfsRevalidate(file->f_dentry);
if (result) {
LOG(4, (KERN_DEBUG "VMware hgfs: HgfsMmap: invalid dentry\n"));
goto out;
}
result = generic_file_mmap(file, vma);
out:
return result;
}
static int
v9fs_mmap_file_mmap(struct file *filp, struct vm_area_struct *vma)
{
int retval;
struct inode *inode;
struct v9fs_inode *v9inode;
struct p9_fid *fid;
inode = file_inode(filp);
v9inode = V9FS_I(inode);
mutex_lock(&v9inode->v_mutex);
if (!v9inode->writeback_fid &&
(vma->vm_flags & VM_WRITE)) {
/*
* clone a fid and add it to writeback_fid
* we do it during mmap instead of
* page dirty time via write_begin/page_mkwrite
* because we want write after unlink usecase
* to work.
*/
fid = v9fs_writeback_fid(file_dentry(filp));
if (IS_ERR(fid)) {
retval = PTR_ERR(fid);
mutex_unlock(&v9inode->v_mutex);
return retval;
}
v9inode->writeback_fid = (void *) fid;
}
mutex_unlock(&v9inode->v_mutex);
retval = generic_file_mmap(filp, vma);
if (!retval)
vma->vm_ops = &v9fs_mmap_file_vm_ops;
return retval;
}
static int ccfs_file_mmap(struct file * file, struct vm_area_struct * vma) {
mdbg(INFO3,"Mmap file %p [%s]", file, file->f_dentry->d_name.name);
return generic_file_mmap(file, vma);
}
static int unionfs_mmap(struct file *file, struct vm_area_struct *vma)
{
int err = 0;
bool willwrite;
struct file *lower_file;
struct dentry *dentry = file->f_path.dentry;
struct dentry *parent;
struct vm_operations_struct *saved_vm_ops = NULL;
unionfs_read_lock(dentry->d_sb, UNIONFS_SMUTEX_PARENT);
parent = unionfs_lock_parent(dentry, UNIONFS_DMUTEX_PARENT);
unionfs_lock_dentry(dentry, UNIONFS_DMUTEX_CHILD);
/* This might be deferred to mmap's writepage */
willwrite = ((vma->vm_flags | VM_SHARED | VM_WRITE) == vma->vm_flags);
err = unionfs_file_revalidate(file, parent, willwrite);
if (unlikely(err))
goto out;
unionfs_check_file(file);
/*
* File systems which do not implement ->writepage may use
* generic_file_readonly_mmap as their ->mmap op. If you call
* generic_file_readonly_mmap with VM_WRITE, you'd get an -EINVAL.
* But we cannot call the lower ->mmap op, so we can't tell that
* writeable mappings won't work. Therefore, our only choice is to
* check if the lower file system supports the ->writepage, and if
* not, return EINVAL (the same error that
* generic_file_readonly_mmap returns in that case).
*/
lower_file = unionfs_lower_file(file);
if (willwrite && !lower_file->f_mapping->a_ops->writepage) {
err = -EINVAL;
printk(KERN_ERR "unionfs: branch %d file system does not "
"support writeable mmap\n", fbstart(file));
goto out;
}
/*
* find and save lower vm_ops.
*
* XXX: the VFS should have a cleaner way of finding the lower vm_ops
*/
if (!UNIONFS_F(file)->lower_vm_ops) {
err = lower_file->f_op->mmap(lower_file, vma);
if (err) {
printk(KERN_ERR "unionfs: lower mmap failed %d\n", err);
goto out;
}
saved_vm_ops = vma->vm_ops;
err = do_munmap(current->mm, vma->vm_start,
vma->vm_end - vma->vm_start);
if (err) {
printk(KERN_ERR "unionfs: do_munmap failed %d\n", err);
goto out;
}
}
file->f_mapping->a_ops = &unionfs_dummy_aops;
err = generic_file_mmap(file, vma);
file->f_mapping->a_ops = &unionfs_aops;
if (err) {
printk(KERN_ERR "unionfs: generic_file_mmap failed %d\n", err);
goto out;
}
vma->vm_ops = &unionfs_vm_ops;
if (!UNIONFS_F(file)->lower_vm_ops)
UNIONFS_F(file)->lower_vm_ops = saved_vm_ops;
out:
if (!err) {
/* copyup could cause parent dir times to change */
unionfs_copy_attr_times(parent->d_inode);
unionfs_check_file(file);
}
unionfs_unlock_dentry(dentry);
unionfs_unlock_parent(dentry, parent);
unionfs_read_unlock(dentry->d_sb);
return err;
}
int sjfs_fops_mmap(struct file *f, struct vm_area_struct *v) {
printk("sjfs_fops_mmap -> generic_file_mmap\n");
return generic_file_mmap(f, v);
}
/*
* @breif called by the mmap(2) system call ? dont understance...
*
* @param pfile file structure to map on
* @param vma virtual memory address?
*
* returns error codes
*/
int yramfs_file_mmap(struct file * pfile, struct vm_area_struct * vma)
{
DBG_PRINT("mmap");
return generic_file_mmap(pfile, vma);
}
