/*
* A single inode exist for all anon_inode files. Contrary to pipes,
* anon_inode inodes has no per-instance data associated, so we can avoid
* the allocation of multiple of them.
*/
static struct inode *anon_inode_mkinode(void)
{
struct inode *inode = new_inode(anon_inode_mnt->mnt_sb);
struct _inode *_inode;
if (!inode)
return ERR_PTR(-ENOMEM);
_inode = tx_cache_get_inode(inode);
_inode->i_fop = &anon_inode_fops;
/*
* Mark the inode dirty from the very beginning,
* that way it will never be moved to the dirty
* list because mark_inode_dirty() will think
* that it already _is_ on the dirty list.
*/
inode->i_state = I_DIRTY;
_inode->i_mode = S_IRUSR | S_IWUSR;
_inode->i_uid = current->fsuid;
_inode->i_gid = current->fsgid;
_inode->i_atime = _inode->i_mtime = _inode->i_ctime = CURRENT_TIME;
return inode;
}
int abort_dentry(struct txobj_thread_list_node * xnode){
struct dentry * orig = (struct dentry *)xnode->orig_obj;
struct _dentry * shadow = (struct _dentry *)xnode->shadow_obj;
if(xnode->rw == ACCESS_R){
atomic_dec(&shadow->tx_readcount);
atomic_dec(&shadow->tx_refcount);
} else {
if((shadow->d_flags & DCACHE_SPECULATIVE_CREATE)){
struct _dentry *old_dentry = orig->d_contents;
/* Free up the speculatively created inode reference to force deletion */
tx_atomic_dec(&shadow->d_inode->i_count);
tx_cache_get_inode(shadow->d_inode)->i_nlink--;
free_tx_dentry(shadow);
/* Use this as a signal to release_dentry */
old_dentry->d_flags |= DCACHE_SPECULATIVE_CREATE;
/* We also need the dentry to be negative */
KSTM_BUG_ON(old_dentry->d_inode != NULL);
/* And hack to avoid messing up the parent
* dentry's refcount when we put, as we will
* rollback the refcount increase elsewhere
*/
tx_atomic_inc(&old_dentry->d_parent->d_count);
} else
free_tx_dentry(shadow);
}
return 0;
}
static int oprofilefs_fill_super(struct super_block * sb, void * data, int silent)
{
struct inode * root_inode;
struct _inode * _root_inode;
struct dentry * root_dentry;
sb->s_blocksize = PAGE_CACHE_SIZE;
sb->s_blocksize_bits = PAGE_CACHE_SHIFT;
sb->s_magic = OPROFILEFS_MAGIC;
sb->s_op = &s_ops;
sb->s_time_gran = 1;
root_inode = oprofilefs_get_inode(sb, S_IFDIR | 0755);
if (!root_inode)
return -ENOMEM;
_root_inode = tx_cache_get_inode(root_inode);
_root_inode->i_op = &simple_dir_inode_operations;
_root_inode->i_fop = &simple_dir_operations;
root_dentry = d_alloc_root(_root_inode);
if (!root_dentry) {
iput(root_inode);
return -ENOMEM;
}
sb->s_root = root_dentry;
oprofile_create_files(sb, tx_cache_get_dentry(root_dentry));
// FIXME: verify kill_litter_super removes our dentries
return 0;
}
void swap_unplug_io_fn(struct backing_dev_info *unused_bdi, struct page *page)
{
swp_entry_t entry;
down_read(&swap_unplug_sem);
entry.val = page_private(page);
if (PageSwapCache(page)) {
struct block_device *bdev = swap_info[swp_type(entry)].bdev;
struct backing_dev_info *bdi;
/*
* If the page is removed from swapcache from under us (with a
* racy try_to_unuse/swapoff) we need an additional reference
* count to avoid reading garbage from page_private(page) above.
* If the WARN_ON triggers during a swapoff it maybe the race
* condition and it's harmless. However if it triggers without
* swapoff it signals a problem.
*/
WARN_ON(page_count(page) <= 1);
bdi = tx_cache_get_inode(bdev->bd_inode)->i_mapping->backing_dev_info;
blk_run_backing_dev(bdi, page);
}
up_read(&swap_unplug_sem);
}
static struct inode * oprofilefs_get_inode(struct super_block * sb, int mode)
{
struct inode * inode = new_inode(sb);
if (inode) {
struct _inode *_inode = tx_cache_get_inode(inode);
_inode->i_mode = mode;
_inode->i_uid = 0;
_inode->i_gid = 0;
inode->i_blocks = 0;
_inode->i_atime = _inode->i_mtime = _inode->i_ctime = CURRENT_TIME;
}
return inode;
}
static struct _dentry * __oprofilefs_create_file(struct super_block * sb,
struct _dentry * root, char const * name, const struct file_operations * fops,
int perm)
{
struct dentry * dentry;
struct _dentry * _dentry;
struct inode * inode;
struct _inode * _inode;
dentry = d_alloc_name(root, name);
if (!dentry)
return NULL;
inode = oprofilefs_get_inode(sb, S_IFREG | perm);
if (!inode) {
dput(dentry);
return NULL;
}
_inode = tx_cache_get_inode(inode);
_inode->i_fop = fops;
_dentry = tx_cache_get_dentry(dentry);
d_add(_dentry, _inode);
return _dentry;
}
struct _dentry * oprofilefs_mkdir(struct super_block * sb,
struct _dentry * root, char const * name)
{
struct dentry * dentry;
struct _dentry * _dentry;
struct inode * inode;
struct _inode * _inode;
dentry = d_alloc_name(root, name);
if (!dentry)
return NULL;
inode = oprofilefs_get_inode(sb, S_IFDIR | 0755);
if (!inode) {
dput(dentry);
return NULL;
}
_inode = tx_cache_get_inode(inode);
_inode->i_op = &simple_dir_inode_operations;
_inode->i_fop = &simple_dir_operations;
_dentry = tx_cache_get_dentry(dentry);
d_add(_dentry, _inode);
return _dentry;
}
static struct file *__dentry_open(struct _dentry *dentry, struct vfsmount *mnt,
int flags, struct file *f,
int (*open)(struct _inode *, struct file *))
{
struct _inode *inode;
int error;
struct super_block *sb;
struct _file *_f = tx_cache_get_file(f);
_f->f_flags = flags;
_f->f_mode = ((flags+1) & O_ACCMODE) | FMODE_LSEEK |
FMODE_PREAD | FMODE_PWRITE;
inode = d_get_inode(dentry);
sb = inode->i_sb;
if (_f->f_mode & FMODE_WRITE) {
error = get_write_access(parent(inode));
if (error)
goto cleanup_file;
}
f->f_mapping = inode->i_mapping;
_f->f_path.dentry = parent(dentry);
_f->f_path.mnt = mnt;
_f->f_pos = 0;
f->f_op = fops_get(inode->i_fop);
file_move(f, &sb->s_files);
if (!open && f->f_op)
open = f->f_op->open;
if (open) {
error = open(inode, f);
if (error)
goto cleanup_all;
}
_f->f_flags &= ~(O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC);
file_ra_state_init(&f->f_ra,
tx_cache_get_inode(f->f_mapping->host)->i_mapping);
/* NB: we're sure to have correct a_ops only after f_op->open */
if (_f->f_flags & O_DIRECT) {
if (!f->f_mapping->a_ops ||
((!f->f_mapping->a_ops->direct_IO) &&
(!f->f_mapping->a_ops->get_xip_page))) {
fput(f);
f = ERR_PTR(-EINVAL);
}
}
return f;
cleanup_all:
fops_put(f->f_op);
if (_f->f_mode & FMODE_WRITE)
put_write_access(parent(inode));
file_kill(f);
_f->f_path.dentry = NULL;
_f->f_path.mnt = NULL;
cleanup_file:
/* Avoid issues if we recycle this object */
if(live_transaction())
early_release(&f->xobj, 1);
put_filp(f);
dput(parent(dentry));
mntput(mnt);
return ERR_PTR(error);
}
/*
* Code shared between mknod, mkdir, symlink and link
*/
static int create_new_entry(struct fuse_conn *fc, struct fuse_req *req,
struct _inode *dir, struct dentry *entry,
int mode)
{
struct fuse_entry_out outarg;
struct inode *inode;
struct _inode *_inode;
int err;
struct fuse_req *forget_req;
forget_req = fuse_get_req(fc);
if (IS_ERR(forget_req)) {
fuse_put_request(fc, req);
return PTR_ERR(forget_req);
}
req->in.h.nodeid = _get_node_id(dir);
req->out.numargs = 1;
req->out.args[0].size = sizeof(outarg);
req->out.args[0].value = &outarg;
request_send(fc, req);
err = req->out.h.error;
fuse_put_request(fc, req);
if (err)
goto out_put_forget_req;
err = -EIO;
if (invalid_nodeid(outarg.nodeid))
goto out_put_forget_req;
if ((outarg.attr.mode ^ mode) & S_IFMT)
goto out_put_forget_req;
inode = fuse_iget(dir->i_sb, outarg.nodeid, outarg.generation,
&outarg.attr);
if (!inode) {
fuse_send_forget(fc, forget_req, outarg.nodeid, 1);
return -ENOMEM;
}
fuse_put_request(fc, forget_req);
_inode = tx_cache_get_inode(inode);
if (S_ISDIR(_inode->i_mode)) {
struct dentry *alias;
mutex_lock(&fc->inst_mutex);
alias = d_find_alias(_inode);
if (alias) {
/* New directory must have moved since mkdir */
mutex_unlock(&fc->inst_mutex);
dput(alias);
iput(inode);
return -EBUSY;
}
d_instantiate(entry, _inode);
mutex_unlock(&fc->inst_mutex);
} else
d_instantiate(entry, _inode);
fuse_change_timeout(entry, &outarg);
fuse_invalidate_attr(dir);
return 0;
out_put_forget_req:
fuse_put_request(fc, forget_req);
return err;
}
/*
* Atomic create+open operation
*
* If the filesystem doesn't support this, then fall back to separate
* 'mknod' + 'open' requests.
*/
static int fuse_create_open(struct _inode *dir, struct dentry *entry, int mode,
struct nameidata *nd)
{
int err;
struct inode *inode;
struct _inode *_inode;
struct fuse_conn *fc = get_fuse_conn(dir);
struct fuse_req *req;
struct fuse_req *forget_req;
struct fuse_open_in inarg;
struct fuse_open_out outopen;
struct fuse_entry_out outentry;
struct fuse_file *ff;
struct file *file;
int flags = nd->intent.open.flags - 1;
if (fc->no_create)
return -ENOSYS;
forget_req = fuse_get_req(fc);
if (IS_ERR(forget_req))
return PTR_ERR(forget_req);
req = fuse_get_req(fc);
err = PTR_ERR(req);
if (IS_ERR(req))
goto out_put_forget_req;
err = -ENOMEM;
ff = fuse_file_alloc();
if (!ff)
goto out_put_request;
flags &= ~O_NOCTTY;
memset(&inarg, 0, sizeof(inarg));
inarg.flags = flags;
inarg.mode = mode;
req->in.h.opcode = FUSE_CREATE;
req->in.h.nodeid = _get_node_id(dir);
req->in.numargs = 2;
req->in.args[0].size = sizeof(inarg);
req->in.args[0].value = &inarg;
req->in.args[1].size = entry->d_name.len + 1;
req->in.args[1].value = entry->d_name.name;
req->out.numargs = 2;
req->out.args[0].size = sizeof(outentry);
req->out.args[0].value = &outentry;
req->out.args[1].size = sizeof(outopen);
req->out.args[1].value = &outopen;
request_send(fc, req);
err = req->out.h.error;
if (err) {
if (err == -ENOSYS)
fc->no_create = 1;
goto out_free_ff;
}
err = -EIO;
if (!S_ISREG(outentry.attr.mode) || invalid_nodeid(outentry.nodeid))
goto out_free_ff;
fuse_put_request(fc, req);
inode = fuse_iget(dir->i_sb, outentry.nodeid, outentry.generation,
&outentry.attr);
if (!inode) {
flags &= ~(O_CREAT | O_EXCL | O_TRUNC);
ff->fh = outopen.fh;
fuse_sync_release(fc, ff, outentry.nodeid, flags);
fuse_send_forget(fc, forget_req, outentry.nodeid, 1);
return -ENOMEM;
}
_inode = tx_cache_get_inode(inode);
fuse_put_request(fc, forget_req);
d_instantiate(entry, _inode);
fuse_change_timeout(entry, &outentry);
file = lookup_instantiate_filp(nd, entry, generic_file_open);
if (IS_ERR(file)) {
ff->fh = outopen.fh;
fuse_sync_release(fc, ff, outentry.nodeid, flags);
return PTR_ERR(file);
}
fuse_finish_open(_inode, file, ff, &outopen);
return 0;
out_free_ff:
fuse_file_free(ff);
out_put_request:
fuse_put_request(fc, req);
out_put_forget_req:
fuse_put_request(fc, forget_req);
return err;
}
static struct dentry *fuse_lookup(struct _inode *dir, struct dentry *entry,
struct nameidata *nd)
{
int err;
struct fuse_entry_out outarg;
struct inode *inode = NULL;
struct _inode *_inode = NULL;
struct fuse_conn *fc = get_fuse_conn(dir);
struct fuse_req *req;
struct fuse_req *forget_req;
if (entry->d_name.len > FUSE_NAME_MAX)
return ERR_PTR(-ENAMETOOLONG);
req = fuse_get_req(fc);
if (IS_ERR(req))
return ERR_PTR(PTR_ERR(req));
forget_req = fuse_get_req(fc);
if (IS_ERR(forget_req)) {
fuse_put_request(fc, req);
return ERR_PTR(PTR_ERR(forget_req));
}
fuse_lookup_init(req, dir, entry, &outarg);
request_send(fc, req);
err = req->out.h.error;
fuse_put_request(fc, req);
/* Zero nodeid is same as -ENOENT, but with valid timeout */
if (!err && outarg.nodeid &&
(invalid_nodeid(outarg.nodeid) ||
!fuse_valid_type(outarg.attr.mode)))
err = -EIO;
if (!err && outarg.nodeid) {
inode = fuse_iget(dir->i_sb, outarg.nodeid, outarg.generation,
&outarg.attr);
if (!inode) {
fuse_send_forget(fc, forget_req, outarg.nodeid, 1);
return ERR_PTR(-ENOMEM);
}
}
fuse_put_request(fc, forget_req);
if (err && err != -ENOENT)
return ERR_PTR(err);
_inode = tx_cache_get_inode(inode);
if (inode && S_ISDIR(_inode->i_mode)) {
mutex_lock(&fc->inst_mutex);
err = fuse_d_add_directory(entry, _inode);
mutex_unlock(&fc->inst_mutex);
if (err) {
iput(inode);
return ERR_PTR(err);
}
} else
d_add(entry, _inode);
entry->d_op = &fuse_dentry_operations;
if (!err)
fuse_change_timeout(entry, &outarg);
else
fuse_invalidate_entry_cache(entry);
return NULL;
}
/**
* anon_inode_getfd - creates a new file instance by hooking it up to and
* anonymous inode, and a dentry that describe the "class"
* of the file
*
* @pfd: [out] pointer to the file descriptor
* @dpinode: [out] pointer to the inode
* @pfile: [out] pointer to the file struct
* @name: [in] name of the "class" of the new file
* @fops [in] file operations for the new file
* @priv [in] private data for the new file (will be file's private_data)
*
* Creates a new file by hooking it on a single inode. This is useful for files
* that do not need to have a full-fledged inode in order to operate correctly.
* All the files created with anon_inode_getfd() will share a single inode, by
* hence saving memory and avoiding code duplication for the file/inode/dentry
* setup.
*/
int anon_inode_getfd(int *pfd, struct inode **pinode, struct file **pfile,
const char *name, const struct file_operations *fops,
void *priv)
{
struct qstr this;
struct dentry *dentry;
struct _dentry *_dentry;
struct inode *inode;
struct _inode *_inode;
struct file *file;
struct _file *_file;
int error, fd;
if (IS_ERR(anon_inode_inode))
return -ENODEV;
file = get_empty_filp();
if (!file)
return -ENFILE;
_file = tx_cache_get_file(file);
inode = igrab(anon_inode_inode);
if (IS_ERR(inode)) {
error = PTR_ERR(inode);
goto err_put_filp;
}
_inode = tx_cache_get_inode(inode);
error = get_unused_fd();
if (error < 0)
goto err_iput;
fd = error;
/*
* Link the inode to a directory entry by creating a unique name
* using the inode sequence number.
*/
error = -ENOMEM;
this.name = name;
this.len = strlen(name);
this.hash = 0;
dentry = d_alloc(tx_cache_get_dentry(anon_inode_mnt->mnt_sb->s_root), &this);
if (!dentry)
goto err_put_unused_fd;
_dentry = tx_cache_get_dentry(dentry);
_dentry->d_op = &anon_inodefs_dentry_operations;
/* Do not publish this dentry inside the global dentry hash table */
_dentry->d_flags &= ~DCACHE_UNHASHED;
d_instantiate(_dentry, _inode);
_file->f_path.mnt = mntget(anon_inode_mnt);
_file->f_path.dentry = dentry;
file->f_mapping = _inode->i_mapping;
_file->f_pos = 0;
_file->f_flags = O_RDWR;
file->f_op = fops;
_file->f_mode = FMODE_READ | FMODE_WRITE;
_file->f_version = 0;
file->private_data = priv;
fd_install(fd, file);
*pfd = fd;
*pinode = inode;
*pfile = file;
return 0;
err_put_unused_fd:
put_unused_fd(fd);
err_iput:
iput(inode);
err_put_filp:
put_filp(file);
return error;
}
/*
* This is the worker routine which does all the work of mapping the disk
* blocks and constructs largest possible bios, submits them for IO if the
* blocks are not contiguous on the disk.
*
* We pass a buffer_head back and forth and use its buffer_mapped() flag to
* represent the validity of its disk mapping and to decide when to do the next
* get_block() call.
*/
static struct bio *
do_mpage_readpage(struct bio *bio, struct page *page, unsigned nr_pages,
sector_t *last_block_in_bio, struct buffer_head *map_bh,
unsigned long *first_logical_block, get_block_t get_block)
{
struct _inode *inode = tx_cache_get_inode(page->mapping->host);
const unsigned blkbits = inode->i_blkbits;
const unsigned blocks_per_page = PAGE_CACHE_SIZE >> blkbits;
const unsigned blocksize = 1 << blkbits;
sector_t block_in_file;
sector_t last_block;
sector_t last_block_in_file;
sector_t blocks[MAX_BUF_PER_PAGE];
unsigned page_block;
unsigned first_hole = blocks_per_page;
struct block_device *bdev = NULL;
int length;
int fully_mapped = 1;
unsigned nblocks;
unsigned relative_block;
if (page_has_buffers(page))
goto confused;
block_in_file = (sector_t)page->index << (PAGE_CACHE_SHIFT - blkbits);
last_block = block_in_file + nr_pages * blocks_per_page;
last_block_in_file = (i_size_read(inode) + blocksize - 1) >> blkbits;
if (last_block > last_block_in_file)
last_block = last_block_in_file;
page_block = 0;
/*
* Map blocks using the result from the previous get_blocks call first.
*/
nblocks = map_bh->b_size >> blkbits;
if (buffer_mapped(map_bh) && block_in_file > *first_logical_block &&
block_in_file < (*first_logical_block + nblocks)) {
unsigned map_offset = block_in_file - *first_logical_block;
unsigned last = nblocks - map_offset;
for (relative_block = 0; ; relative_block++) {
if (relative_block == last) {
clear_buffer_mapped(map_bh);
break;
}
if (page_block == blocks_per_page)
break;
blocks[page_block] = map_bh->b_blocknr + map_offset +
relative_block;
page_block++;
block_in_file++;
}
bdev = map_bh->b_bdev;
}
/*
* Then do more get_blocks calls until we are done with this page.
*/
map_bh->b_page = page;
while (page_block < blocks_per_page) {
map_bh->b_state = 0;
map_bh->b_size = 0;
if (block_in_file < last_block) {
map_bh->b_size = (last_block-block_in_file) << blkbits;
if (get_block(inode, block_in_file, map_bh, 0))
goto confused;
*first_logical_block = block_in_file;
}
if (!buffer_mapped(map_bh)) {
fully_mapped = 0;
if (first_hole == blocks_per_page)
first_hole = page_block;
page_block++;
block_in_file++;
clear_buffer_mapped(map_bh);
continue;
}
/* some filesystems will copy data into the page during
* the get_block call, in which case we don't want to
* read it again. map_buffer_to_page copies the data
* we just collected from get_block into the page's buffers
* so readpage doesn't have to repeat the get_block call
*/
if (buffer_uptodate(map_bh)) {
map_buffer_to_page(page, map_bh, page_block);
goto confused;
}
if (first_hole != blocks_per_page)
goto confused; /* hole -> non-hole */
/* Contiguous blocks? */
if (page_block && blocks[page_block-1] != map_bh->b_blocknr-1)
goto confused;
nblocks = map_bh->b_size >> blkbits;
for (relative_block = 0; ; relative_block++) {
if (relative_block == nblocks) {
clear_buffer_mapped(map_bh);
break;
} else if (page_block == blocks_per_page)
break;
blocks[page_block] = map_bh->b_blocknr+relative_block;
page_block++;
block_in_file++;
}
bdev = map_bh->b_bdev;
}
if (first_hole != blocks_per_page) {
zero_user_page(page, first_hole << blkbits,
PAGE_CACHE_SIZE - (first_hole << blkbits),
KM_USER0);
if (first_hole == 0) {
SetPageUptodate(page);
unlock_page(page);
goto out;
}
} else if (fully_mapped) {
SetPageMappedToDisk(page);
}
/*
* This page will go to BIO. Do we need to send this BIO off first?
*/
if (bio && (*last_block_in_bio != blocks[0] - 1))
bio = mpage_bio_submit(READ, bio);
alloc_new:
if (bio == NULL) {
bio = mpage_alloc(bdev, blocks[0] << (blkbits - 9),
min_t(int, nr_pages, bio_get_nr_vecs(bdev)),
GFP_KERNEL);
if (bio == NULL)
goto confused;
}