/**
* ntfs_file_fsync - sync a file to disk
* @filp: file to be synced
* @dentry: dentry describing the file to sync
* @datasync: if non-zero only flush user data and not metadata
*
* Data integrity sync of a file to disk. Used for fsync, fdatasync, and msync
* system calls. This function is inspired by fs/buffer.c::file_fsync().
*
* If @datasync is false, write the mft record and all associated extent mft
* records as well as the $DATA attribute and then sync the block device.
*
* If @datasync is true and the attribute is non-resident, we skip the writing
* of the mft record and all associated extent mft records (this might still
* happen due to the write_inode_now() call).
*
* Also, if @datasync is true, we do not wait on the inode to be written out
* but we always wait on the page cache pages to be written out.
*
* Note: In the past @filp could be NULL so we ignore it as we don't need it
* anyway.
*
* Locking: Caller must hold i_sem on the inode.
*
* TODO: We should probably also write all attribute/index inodes associated
* with this inode but since we have no simple way of getting to them we ignore
* this problem for now.
*/
static int ntfs_file_fsync(struct file *filp, struct dentry *dentry,
int datasync)
{
struct inode *vi = dentry->d_inode;
int err, ret = 0;
ntfs_debug("Entering for inode 0x%lx.", vi->i_ino);
BUG_ON(S_ISDIR(vi->i_mode));
if (!datasync || !NInoNonResident(NTFS_I(vi)))
ret = ntfs_write_inode(vi, 1);
write_inode_now(vi, !datasync);
/*
* NOTE: If we were to use mapping->private_list (see ext2 and
* fs/buffer.c) for dirty blocks then we could optimize the below to be
* sync_mapping_buffers(vi->i_mapping).
*/
err = sync_blockdev(vi->i_sb->s_bdev);
if (unlikely(err && !ret))
ret = err;
if (likely(!ret))
ntfs_debug("Done.");
else
ntfs_warning(vi->i_sb, "Failed to f%ssync inode 0x%lx. Error "
"%u.", datasync ? "data" : "", vi->i_ino, -ret);
return ret;
}
static int fat_cont_expand(struct inode *inode, loff_t size)
{
struct address_space *mapping = inode->i_mapping;
loff_t start = inode->i_size, count = size - inode->i_size;
int err;
err = generic_cont_expand_simple(inode, size);
if (err)
goto out;
inode->i_ctime = inode->i_mtime = CURRENT_TIME_SEC;
mark_inode_dirty(inode);
if (IS_SYNC(inode)) {
int err2;
/*
* Opencode syncing since we don't have a file open to use
* standard fsync path.
*/
err = filemap_fdatawrite_range(mapping, start,
start + count - 1);
err2 = sync_mapping_buffers(mapping);
if (!err)
err = err2;
err2 = write_inode_now(inode, 1);
if (!err)
err = err2;
if (!err) {
err = filemap_fdatawait_range(mapping, start,
start + count - 1);
}
}
out:
return err;
}
/* Support function for ocfs2_delete_inode. Will help us keep the
* inode data in a consistent state for clear_inode. Always truncates
* pages, optionally sync's them first. */
static void ocfs2_cleanup_delete_inode(struct inode *inode,
int sync_data)
{
mlog(0, "Cleanup inode %llu, sync = %d\n",
(unsigned long long)OCFS2_I(inode)->ip_blkno, sync_data);
if (sync_data)
write_inode_now(inode, 1);
truncate_inode_pages(&inode->i_data, 0);
}
/* Called under inode_lock, with no more references on the
* struct inode, so it's safe here to check the flags field
* and to manipulate i_nlink without any other locks. */
int ocfs2_drop_inode(struct inode *inode)
{
struct ocfs2_inode_info *oi = OCFS2_I(inode);
trace_ocfs2_drop_inode((unsigned long long)oi->ip_blkno,
inode->i_nlink, oi->ip_flags);
assert_spin_locked(&inode->i_lock);
inode->i_state |= I_WILL_FREE;
spin_unlock(&inode->i_lock);
write_inode_now(inode, 1);
spin_lock(&inode->i_lock);
WARN_ON(inode->i_state & I_NEW);
inode->i_state &= ~I_WILL_FREE;
return 1;
}
/*
* Generic function to fsync a file.
*/
int file_fsync(struct file *filp, int datasync)
{
struct inode *inode = filp->f_mapping->host;
struct super_block * sb;
int ret, err;
/* sync the inode to buffers */
ret = write_inode_now(inode, 0);
/* sync the superblock to buffers */
sb = inode->i_sb;
if (sb->s_dirt && sb->s_op->write_super)
sb->s_op->write_super(sb);
/* .. finally sync the buffers to disk */
err = sync_blockdev(sb->s_bdev);
if (!ret)
ret = err;
return ret;
}
/**
* ntfs_file_fsync - sync a file to disk
* @filp: file to be synced
* @dentry: dentry describing the file to sync
* @datasync: if non-zero only flush user data and not metadata
*
* Data integrity sync of a file to disk. Used for fsync, fdatasync, and msync
* system calls. This function is inspired by fs/buffer.c::file_fsync().
*
* If @datasync is false, write the mft record and all associated extent mft
* records as well as the $DATA attribute and then sync the block device.
*
* If @datasync is true and the attribute is non-resident, we skip the writing
* of the mft record and all associated extent mft records (this might still
* happen due to the write_inode_now() call).
*
* Also, if @datasync is true, we do not wait on the inode to be written out
* but we always wait on the page cache pages to be written out.
*
* Note: In the past @filp could be NULL so we ignore it as we don't need it
* anyway.
*
* Locking: Caller must hold i_sem on the inode.
*
* TODO: We should probably also write all attribute/index inodes associated
* with this inode but since we have no simple way of getting to them we ignore
* this problem for now.
*/
static int ntfs_file_fsync(struct file *filp, struct dentry *dentry,
int datasync)
{
struct inode *vi = dentry->d_inode;
int err, ret = 0;
ntfs_debug("Entering for inode 0x%lx.", vi->i_ino);
BUG_ON(S_ISDIR(vi->i_mode));
if (!datasync || !NInoNonResident(NTFS_I(vi)))
ret = ntfs_write_inode(vi, 1);
write_inode_now(vi, !datasync);
err = sync_blockdev(vi->i_sb->s_bdev);
if (unlikely(err && !ret))
ret = err;
if (likely(!ret))
ntfs_debug("Done.");
else
ntfs_warning(vi->i_sb, "Failed to f%ssync inode 0x%lx. Error "
"%u.", datasync ? "data" : "", vi->i_ino, -ret);
return ret;
}
static int gfs2_fsync(struct file *file, struct dentry *dentry, int datasync)
{
struct inode *inode = dentry->d_inode;
int sync_state = inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC);
int ret = 0;
if (gfs2_is_jdata(GFS2_I(inode))) {
gfs2_log_flush(GFS2_SB(inode), GFS2_I(inode)->i_gl);
return 0;
}
if (sync_state != 0) {
if (!datasync)
ret = write_inode_now(inode, 0);
if (gfs2_is_stuffed(GFS2_I(inode)))
gfs2_log_flush(GFS2_SB(inode), GFS2_I(inode)->i_gl);
}
return ret;
}
static int exofs_file_fsync(struct file *filp, int datasync)
{
int ret;
struct address_space *mapping = filp->f_mapping;
struct inode *inode = mapping->host;
struct super_block *sb;
ret = filemap_write_and_wait(mapping);
if (ret)
return ret;
/* sync the inode attributes */
ret = write_inode_now(inode, 1);
/* This is a good place to write the sb */
/* TODO: Sechedule an sb-sync on create */
sb = inode->i_sb;
if (sb->s_dirt)
exofs_sync_fs(sb, 1);
return ret;
}
/*
* Generic function to fsync a file.
*
* filp may be NULL if called via the msync of a vma.
*/
int file_fsync(struct file *filp, struct dentry *dentry, int datasync)
{
struct inode * inode = dentry->d_inode;
struct super_block * sb;
int ret, err;
/* sync the inode to buffers */
ret = write_inode_now(inode, 0);
/* sync the superblock to buffers */
sb = inode->i_sb;
lock_super(sb);
if (is_sb_dirty(sb) && sb->s_op->write_super)
sb->s_op->write_super(sb);
unlock_super(sb);
/* .. finally sync the buffers to disk */
err = sync_blockdev(sb->s_bdev);
if (!ret)
ret = err;
return ret;
}
ssize_t GSFS_file_write(struct file *filp, const char __user *buf, size_t len, loff_t *off){
struct inode *inode=filp->f_mapping->host;
struct address_space *mapping=filp->f_mapping;
sector_t sec_start,
sec_end,
sec_len;
unsigned long bufstart,
bytes_in_first_buf_page,
bytes_in_last_buf_page,
pagestartbyte,
pageendbyte,
pagelen;
size_t rlen;
struct page *res[2],
*page,
**pages;
unsigned int i,
j,
pages_count,
start_read,
end_read;
gwf(printk("<0>" "File write with inode :%lu len:%lu offset:%llu , filepos:%llu, buf:%lx inode_size:%llu, pid:%u\n",inode->i_ino,len,*off,filp->f_pos,(unsigned long)buf,inode->i_size,current->pid));
if(unlikely(!access_ok(VERIFY_READ,buf,len)))
return -1;
mutex_lock(&inode->i_mutex);
if((*off+len)>inode->i_size){
if((*off+len)>inode->i_sb->s_maxbytes)
return -1;
inode->i_size=(*off+len);
GSFS_truncate(inode);
}
if(filp->f_flags & O_APPEND)
*off=inode->i_size;
current->backing_dev_info=mapping->backing_dev_info;
file_remove_suid(filp);
file_update_time(filp);
//inode_inc_iversion(inode);
sec_start=(*off)>>Block_Size_Bits;
sec_end=(*off+len-1)>>Block_Size_Bits;
sec_len=sec_end-sec_start+1;
pages=kzalloc(sizeof(struct page*) * sec_len, GFP_KERNEL);
bytes_in_first_buf_page=Block_Size-((*off)&((unsigned long)Block_Size-1));
bytes_in_last_buf_page=((*off+len)&((unsigned long)Block_Size-1));
if(bytes_in_last_buf_page==0)
bytes_in_last_buf_page=Block_Size;
start_read=(bytes_in_first_buf_page!=Block_Size)?1:0;
end_read=(bytes_in_last_buf_page!=Block_Size && inode->i_size>(*off+len))?1:0;
gwf(printk("<0>" "GSFS write bytes_in_first_buf_page:%lu, bytes_in_last_buf_page:%lu\n",bytes_in_first_buf_page,bytes_in_last_buf_page));
gwf(printk("<0>" "GSFS write start_read:%u, end_read:%u, sec_start:%lu, sec_end:%lu\n",start_read,end_read,sec_start,sec_end));
if(sec_start==sec_end){
if(start_read || end_read){
res[0]=GSFS_get_data_page_of_inode_with_read(inode, sec_start);
gwf(printk("<0>" "sec_start==sec_end, start_read || end_read , res[0]=%lx",(unsigned long)res[0]));
}
else{
res[0]=GSFS_get_locked_data_page_of_inode_without_read(inode,sec_start);
if(likely(res[0]))
unlock_page(res[0]);
gwf(printk("<0>" "sec_start==sec_end, !(start_read || end_read) , res[0]=%lx",(unsigned long)res[0]));
}
res[1]=0;
if(unlikely(!res[0])){
gwf(printk("<0>" "GSFS write len:-1\n"));
mutex_unlock(&inode->i_mutex);
printk("<1>" "GSFS write len:-1\n");
kfree(pages);
return len;
}
}
else{
if(start_read){
res[0]=GSFS_get_data_page_of_inode_with_read(inode, sec_start);
gwf(printk("<0>" "sec_start!=sec_end, start_read, res[0]=%lx",(unsigned long)res[0]));
}
else{
res[0]=GSFS_get_locked_data_page_of_inode_without_read(inode,sec_start);
if(likely(res[0]))
unlock_page(res[0]);
gwf(printk("<0>" "sec_start!=sec_end, !start_read, res[0]=%lx",(unsigned long)res[0]));
}
}
pages_count=0;
if(sec_len>1)
for(i=sec_start+1;i<=sec_end-1;i++)
pages[pages_count++]=GSFS_get_locked_data_page_of_inode_without_read(inode,i);
if(sec_start != sec_end){
if(end_read){
res[1]=GSFS_get_data_page_of_inode_with_read(inode,sec_end);
gwf(printk("<0>" "sec_start!=sec_end, end_read, res[1]=%lx",(unsigned long)res[1]));
}
else{
res[1]=GSFS_get_locked_data_page_of_inode_without_read(inode,sec_end);
if(likely(res[1]))
unlock_page(res[1]);
gwf(printk("<0>" "sec_start!=sec_end, !end_read, res[1]=%lx",(unsigned long)res[1]));
}
if(unlikely(!res[0] || !res[1])){
gwf(printk("<0>" "GSFS write len:-1\n"));
printk("<1>" "GSFS write len:-1\n");
mutex_unlock(&inode->i_mutex);
kfree(pages);
return len;
}
}
rlen=0;
bufstart=(unsigned long)buf+bytes_in_first_buf_page;
pagelen=Block_Size;
//100% expected complete pages that should be copied
pages_count=0;
if(sec_len>1)
for(i=sec_start+1;i<=sec_end-1;i++){
gwf(printk("<0>" "write page complete pages, i:%u, bufstart:%lx, rlen=%lu\n",i,bufstart,rlen));
page=pages[pages_count++];
if(unlikely(!page))
goto buf_cont;
j=__copy_from_user_inatomic(page_address(page),(void*)bufstart,pagelen);
rlen+=(Block_Size-j);
mark_page_accessed(page);
set_page_dirty(page);
put_page(page);
unlock_page(page);
buf_cont:
bufstart+=pagelen;
}
//first and last page that are not surely complete
for(i=0;i<2 && res[i];i++){
page=res[i];
wait_on_page_locked(page);
lock_page(page);
if(page->index==sec_start){
bufstart=(unsigned long)buf;
pagestartbyte=Block_Size-bytes_in_first_buf_page;
if(sec_start==sec_end)
pageendbyte=pagestartbyte+len-1;
else
pageendbyte=Block_Size-1;
}
else{
bufstart=(unsigned long)buf+bytes_in_first_buf_page+((sec_len-2)<<Block_Size_Bits);
pageendbyte=bytes_in_last_buf_page-1;
pagestartbyte=0;
}
gwf(printk("<0>" "gsfs_write for first and last page, i=%u, page:%lx, bufstart:%lx, pagestartbyte:%lu, pageendbyte:%lu\n",
i,(unsigned long)page,bufstart,pagestartbyte,pageendbyte));
pagelen=pageendbyte-pagestartbyte+1;
j=__copy_from_user_inatomic(page_address(page)+pagestartbyte,(void*)bufstart,pagelen);
rlen+=(pagelen-j);
mark_page_accessed(page);
set_page_dirty(page);
put_page(page);
unlock_page(page);
}
mutex_unlock(&inode->i_mutex);
(*off)+=rlen;
gwf(printk("<0>" "GSFS write rlen:%lu\n",rlen));
kfree(pages);
if(filp->f_flags & O_SYNC){
write_inode_now(inode,1);
}
return rlen;
}
/*
* Set various file attributes.
* N.B. After this call fhp needs an fh_put
*/
__be32
nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
int check_guard, time_t guardtime)
{
struct dentry *dentry;
struct inode *inode;
int accmode = NFSD_MAY_SATTR;
int ftype = 0;
__be32 err;
int host_err;
int size_change = 0;
if (iap->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_SIZE))
accmode |= NFSD_MAY_WRITE|NFSD_MAY_OWNER_OVERRIDE;
if (iap->ia_valid & ATTR_SIZE)
ftype = S_IFREG;
/* Get inode */
err = fh_verify(rqstp, fhp, ftype, accmode);
if (err)
goto out;
dentry = fhp->fh_dentry;
inode = dentry->d_inode;
/* Ignore any mode updates on symlinks */
if (S_ISLNK(inode->i_mode))
iap->ia_valid &= ~ATTR_MODE;
if (!iap->ia_valid)
goto out;
/*
* NFSv2 does not differentiate between "set-[ac]time-to-now"
* which only requires access, and "set-[ac]time-to-X" which
* requires ownership.
* So if it looks like it might be "set both to the same time which
* is close to now", and if inode_change_ok fails, then we
* convert to "set to now" instead of "set to explicit time"
*
* We only call inode_change_ok as the last test as technically
* it is not an interface that we should be using. It is only
* valid if the filesystem does not define it's own i_op->setattr.
*/
#define BOTH_TIME_SET (ATTR_ATIME_SET | ATTR_MTIME_SET)
#define MAX_TOUCH_TIME_ERROR (30*60)
if ((iap->ia_valid & BOTH_TIME_SET) == BOTH_TIME_SET &&
iap->ia_mtime.tv_sec == iap->ia_atime.tv_sec) {
/*
* Looks probable.
*
* Now just make sure time is in the right ballpark.
* Solaris, at least, doesn't seem to care what the time
* request is. We require it be within 30 minutes of now.
*/
time_t delta = iap->ia_atime.tv_sec - get_seconds();
if (delta < 0)
delta = -delta;
if (delta < MAX_TOUCH_TIME_ERROR &&
inode_change_ok(inode, iap) != 0) {
/*
* Turn off ATTR_[AM]TIME_SET but leave ATTR_[AM]TIME.
* This will cause notify_change to set these times
* to "now"
*/
iap->ia_valid &= ~BOTH_TIME_SET;
}
}
/*
* The size case is special.
* It changes the file as well as the attributes.
*/
if (iap->ia_valid & ATTR_SIZE) {
if (iap->ia_size < inode->i_size) {
err = nfsd_permission(rqstp, fhp->fh_export, dentry,
NFSD_MAY_TRUNC|NFSD_MAY_OWNER_OVERRIDE);
if (err)
goto out;
}
/*
* If we are changing the size of the file, then
* we need to break all leases.
*/
host_err = break_lease(inode, FMODE_WRITE | O_NONBLOCK);
if (host_err == -EWOULDBLOCK)
host_err = -ETIMEDOUT;
if (host_err) /* ENOMEM or EWOULDBLOCK */
goto out_nfserr;
host_err = get_write_access(inode);
if (host_err)
goto out_nfserr;
size_change = 1;
host_err = locks_verify_truncate(inode, NULL, iap->ia_size);
if (host_err) {
put_write_access(inode);
goto out_nfserr;
}
DQUOT_INIT(inode);
}
/* sanitize the mode change */
if (iap->ia_valid & ATTR_MODE) {
iap->ia_mode &= S_IALLUGO;
iap->ia_mode |= (inode->i_mode & ~S_IALLUGO);
}
/* Revoke setuid/setgid on chown */
if (((iap->ia_valid & ATTR_UID) && iap->ia_uid != inode->i_uid) ||
((iap->ia_valid & ATTR_GID) && iap->ia_gid != inode->i_gid)) {
iap->ia_valid |= ATTR_KILL_PRIV;
if (iap->ia_valid & ATTR_MODE) {
/* we're setting mode too, just clear the s*id bits */
iap->ia_mode &= ~S_ISUID;
if (iap->ia_mode & S_IXGRP)
iap->ia_mode &= ~S_ISGID;
} else {
/* set ATTR_KILL_* bits and let VFS handle it */
iap->ia_valid |= (ATTR_KILL_SUID | ATTR_KILL_SGID);
}
}
/* Change the attributes. */
iap->ia_valid |= ATTR_CTIME;
err = nfserr_notsync;
if (!check_guard || guardtime == inode->i_ctime.tv_sec) {
fh_lock(fhp);
host_err = notify_change(dentry, iap);
err = nfserrno(host_err);
fh_unlock(fhp);
}
if (size_change)
put_write_access(inode);
if (!err)
if (EX_ISSYNC(fhp->fh_export))
write_inode_now(inode, 1);
out:
return err;
out_nfserr:
err = nfserrno(host_err);
goto out;
}
static int setflags(struct inode *inode, int flags)
{
int oldflags, err, release;
struct ubifs_inode *ui = ubifs_inode(inode);
struct ubifs_info *c = inode->i_sb->s_fs_info;
struct ubifs_budget_req req = { .dirtied_ino = 1,
.dirtied_ino_d = ui->data_len };
err = ubifs_budget_space(c, &req);
if (err)
return err;
/*
* The IMMUTABLE and APPEND_ONLY flags can only be changed by
* the relevant capability.
*/
mutex_lock(&ui->ui_mutex);
oldflags = ubifs2ioctl(ui->flags);
if ((flags ^ oldflags) & (FS_APPEND_FL | FS_IMMUTABLE_FL)) {
if (!capable(CAP_LINUX_IMMUTABLE)) {
err = -EPERM;
goto out_unlock;
}
}
ui->flags = ioctl2ubifs(flags);
ubifs_set_inode_flags(inode);
inode->i_ctime = ubifs_current_time(inode);
release = ui->dirty;
mark_inode_dirty_sync(inode);
mutex_unlock(&ui->ui_mutex);
if (release)
ubifs_release_budget(c, &req);
if (IS_SYNC(inode))
err = write_inode_now(inode, 1);
return err;
out_unlock:
ubifs_err("can't modify inode %lu attributes", inode->i_ino);
mutex_unlock(&ui->ui_mutex);
ubifs_release_budget(c, &req);
return err;
}
long ubifs_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
int flags, err;
struct inode *inode = file->f_path.dentry->d_inode;
switch (cmd) {
case FS_IOC_GETFLAGS:
flags = ubifs2ioctl(ubifs_inode(inode)->flags);
dbg_gen("get flags: %#x, i_flags %#x", flags, inode->i_flags);
return put_user(flags, (int __user *) arg);
case FS_IOC_SETFLAGS: {
if (IS_RDONLY(inode))
return -EROFS;
if (!inode_owner_or_capable(inode))
return -EACCES;
if (get_user(flags, (int __user *) arg))
return -EFAULT;
if (!S_ISDIR(inode->i_mode))
flags &= ~FS_DIRSYNC_FL;
/*
* Make sure the file-system is read-write and make sure it
* will not become read-only while we are changing the flags.
*/
err = mnt_want_write(file->f_path.mnt);
if (err)
return err;
dbg_gen("set flags: %#x, i_flags %#x", flags, inode->i_flags);
err = setflags(inode, flags);
mnt_drop_write(file->f_path.mnt);
return err;
}
default:
return -ENOTTY;
}
}
/*
* Set various file attributes.
* N.B. After this call fhp needs an fh_put
*/
int
nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
int check_guard, time_t guardtime)
{
struct dentry *dentry;
struct inode *inode;
int accmode = MAY_SATTR;
int ftype = 0;
int imode;
int err;
int size_change = 0;
if (iap->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_SIZE))
accmode |= MAY_WRITE|MAY_OWNER_OVERRIDE;
if (iap->ia_valid & ATTR_SIZE)
ftype = S_IFREG;
/* Get inode */
err = fh_verify(rqstp, fhp, ftype, accmode);
if (err || !iap->ia_valid)
goto out;
dentry = fhp->fh_dentry;
inode = dentry->d_inode;
/* NFSv2 does not differentiate between "set-[ac]time-to-now"
* which only requires access, and "set-[ac]time-to-X" which
* requires ownership.
* So if it looks like it might be "set both to the same time which
* is close to now", and if inode_change_ok fails, then we
* convert to "set to now" instead of "set to explicit time"
*
* We only call inode_change_ok as the last test as technically
* it is not an interface that we should be using. It is only
* valid if the filesystem does not define it's own i_op->setattr.
*/
#define BOTH_TIME_SET (ATTR_ATIME_SET | ATTR_MTIME_SET)
#define MAX_TOUCH_TIME_ERROR (30*60)
if ((iap->ia_valid & BOTH_TIME_SET) == BOTH_TIME_SET
&& iap->ia_mtime == iap->ia_atime
) {
/* Looks probable. Now just make sure time is in the right ballpark.
* Solaris, at least, doesn't seem to care what the time request is.
* We require it be within 30 minutes of now.
*/
time_t delta = iap->ia_atime - CURRENT_TIME;
if (delta<0) delta = -delta;
if (delta < MAX_TOUCH_TIME_ERROR &&
inode_change_ok(inode, iap) != 0) {
/* turn off ATTR_[AM]TIME_SET but leave ATTR_[AM]TIME
* this will cause notify_change to set these times to "now"
*/
iap->ia_valid &= ~BOTH_TIME_SET;
}
}
/* The size case is special. It changes the file as well as the attributes. */
if (iap->ia_valid & ATTR_SIZE) {
if (iap->ia_size < inode->i_size) {
err = nfsd_permission(fhp->fh_export, dentry, MAY_TRUNC|MAY_OWNER_OVERRIDE);
if (err)
goto out;
}
/*
* If we are changing the size of the file, then
* we need to break all leases.
*/
err = get_lease(inode, FMODE_WRITE);
if (err)
goto out_nfserr;
err = get_write_access(inode);
if (err)
goto out_nfserr;
err = locks_verify_truncate(inode, NULL, iap->ia_size);
if (err) {
put_write_access(inode);
goto out_nfserr;
}
DQUOT_INIT(inode);
}
imode = inode->i_mode;
if (iap->ia_valid & ATTR_MODE) {
iap->ia_mode &= S_IALLUGO;
imode = iap->ia_mode |= (imode & ~S_IALLUGO);
}
/* Revoke setuid/setgid bit on chown/chgrp */
if ((iap->ia_valid & ATTR_UID) && (imode & S_ISUID)
&& iap->ia_uid != inode->i_uid) {
iap->ia_valid |= ATTR_MODE;
iap->ia_mode = imode &= ~S_ISUID;
}
if ((iap->ia_valid & ATTR_GID) && (imode & S_ISGID)
&& iap->ia_gid != inode->i_gid) {
iap->ia_valid |= ATTR_MODE;
iap->ia_mode = imode &= ~S_ISGID;
}
/* Change the attributes. */
iap->ia_valid |= ATTR_CTIME;
if (iap->ia_valid & ATTR_SIZE) {
fh_lock(fhp);
size_change = 1;
}
err = nfserr_notsync;
if (!check_guard || guardtime == inode->i_ctime) {
err = notify_change(dentry, iap);
err = nfserrno(err);
}
if (size_change) {
fh_unlock(fhp);
put_write_access(inode);
}
if (!err)
if (EX_ISSYNC(fhp->fh_export))
write_inode_now(inode, 1);
out:
return err;
out_nfserr:
err = nfserrno(err);
goto out;
}
/*
* Rename a file
* N.B. After this call _both_ ffhp and tfhp need an fh_put
*/
int
nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen,
struct svc_fh *tfhp, char *tname, int tlen)
{
struct dentry *fdentry, *tdentry, *odentry, *ndentry;
struct inode *fdir, *tdir;
int err;
err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_REMOVE);
if (err)
goto out;
err = fh_verify(rqstp, tfhp, S_IFDIR, MAY_CREATE);
if (err)
goto out;
fdentry = ffhp->fh_dentry;
fdir = fdentry->d_inode;
tdentry = tfhp->fh_dentry;
tdir = tdentry->d_inode;
/* N.B. We shouldn't need this ... dentry layer handles it */
err = nfserr_perm;
if (!flen || (fname[0] == '.' &&
(flen == 1 || (flen == 2 && fname[1] == '.'))) ||
!tlen || (tname[0] == '.' &&
(tlen == 1 || (tlen == 2 && tname[1] == '.'))))
goto out;
odentry = lookup_dentry(fname, dget(fdentry), 0);
err = PTR_ERR(odentry);
if (IS_ERR(odentry))
goto out_nfserr;
err = -ENOENT;
if (!odentry->d_inode)
goto out_dput_old;
ndentry = lookup_dentry(tname, dget(tdentry), 0);
err = PTR_ERR(ndentry);
if (IS_ERR(ndentry))
goto out_dput_old;
/*
* Lock the parent directories.
*/
nfsd_double_down(&tdir->i_sem, &fdir->i_sem);
err = -ENOENT;
/* GAM3 check for parent changes after locking. */
if (check_parent(fdir, odentry) &&
check_parent(tdir, ndentry)) {
err = vfs_rename(fdir, odentry, tdir, ndentry);
if (!err && EX_ISSYNC(tfhp->fh_export)) {
write_inode_now(fdir);
write_inode_now(tdir);
}
} else
dprintk("nfsd: Caught race in nfsd_rename");
DQUOT_DROP(fdir);
DQUOT_DROP(tdir);
nfsd_double_up(&tdir->i_sem, &fdir->i_sem);
dput(ndentry);
out_dput_old:
dput(odentry);
if (err)
goto out_nfserr;
out:
return err;
out_nfserr:
err = nfserrno(-err);
goto out;
}
/*
* Unlink a file or directory
* N.B. After this call fhp needs an fh_put
*/
int
nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
char *fname, int flen)
{
struct dentry *dentry, *rdentry;
struct inode *dirp;
int err;
/* N.B. We shouldn't need this test ... handled by dentry layer */
err = nfserr_acces;
if (!flen || isdotent(fname, flen))
goto out;
err = fh_verify(rqstp, fhp, S_IFDIR, MAY_REMOVE);
if (err)
goto out;
dentry = fhp->fh_dentry;
dirp = dentry->d_inode;
rdentry = lookup_dentry(fname, dget(dentry), 0);
err = PTR_ERR(rdentry);
if (IS_ERR(rdentry))
goto out_nfserr;
if (!rdentry->d_inode) {
dput(rdentry);
err = nfserr_noent;
goto out;
}
if (type != S_IFDIR) {
/* It's UNLINK */
err = fh_lock_parent(fhp, rdentry);
if (err)
goto out;
err = vfs_unlink(dirp, rdentry);
DQUOT_DROP(dirp);
fh_unlock(fhp);
dput(rdentry);
} else {
/* It's RMDIR */
/* See comments in fs/namei.c:do_rmdir */
rdentry->d_count++;
nfsd_double_down(&dirp->i_sem, &rdentry->d_inode->i_sem);
if (!fhp->fh_pre_mtime)
fhp->fh_pre_mtime = dirp->i_mtime;
fhp->fh_locked = 1;
err = -ENOENT;
if (check_parent(dirp, rdentry))
err = vfs_rmdir(dirp, rdentry);
rdentry->d_count--;
DQUOT_DROP(dirp);
if (!fhp->fh_post_version)
fhp->fh_post_version = dirp->i_version;
fhp->fh_locked = 0;
nfsd_double_up(&dirp->i_sem, &rdentry->d_inode->i_sem);
dput(rdentry);
}
if (err)
goto out_nfserr;
if (EX_ISSYNC(fhp->fh_export))
write_inode_now(dirp);
out:
return err;
out_nfserr:
err = nfserrno(-err);
goto out;
}
/*
* Set various file attributes.
* N.B. After this call fhp needs an fh_put
*/
int
nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap)
{
struct dentry *dentry;
struct inode *inode;
int accmode = MAY_SATTR;
int ftype = 0;
int imode;
int err;
if (iap->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_SIZE))
accmode |= MAY_WRITE;
if (iap->ia_valid & ATTR_SIZE)
ftype = S_IFREG;
/* Get inode */
err = fh_verify(rqstp, fhp, ftype, accmode);
if (err)
goto out;
dentry = fhp->fh_dentry;
inode = dentry->d_inode;
err = inode_change_ok(inode, iap);
if (err)
goto out_nfserr;
/* The size case is special... */
if (iap->ia_valid & ATTR_SIZE) {
if (!S_ISREG(inode->i_mode))
printk("nfsd_setattr: size change??\n");
if (iap->ia_size < inode->i_size) {
err = nfsd_permission(fhp->fh_export, dentry, MAY_TRUNC);
if (err)
goto out;
}
err = get_write_access(inode);
if (err)
goto out_nfserr;
/* N.B. Should we update the inode cache here? */
inode->i_size = iap->ia_size;
if (inode->i_op && inode->i_op->truncate)
inode->i_op->truncate(inode);
mark_inode_dirty(inode);
put_write_access(inode);
iap->ia_valid &= ~ATTR_SIZE;
iap->ia_valid |= ATTR_MTIME;
iap->ia_mtime = CURRENT_TIME;
}
imode = inode->i_mode;
if (iap->ia_valid & ATTR_MODE) {
iap->ia_mode &= S_IALLUGO;
imode = iap->ia_mode |= (imode & ~S_IALLUGO);
}
/* Revoke setuid/setgid bit on chown/chgrp */
if ((iap->ia_valid & ATTR_UID) && (imode & S_ISUID)
&& iap->ia_uid != inode->i_uid) {
iap->ia_valid |= ATTR_MODE;
iap->ia_mode = imode &= ~S_ISUID;
}
if ((iap->ia_valid & ATTR_GID) && (imode & S_ISGID)
&& iap->ia_gid != inode->i_gid) {
iap->ia_valid |= ATTR_MODE;
iap->ia_mode = imode &= ~S_ISGID;
}
/* Change the attributes. */
if (iap->ia_valid) {
kernel_cap_t saved_cap = 0;
iap->ia_valid |= ATTR_CTIME;
iap->ia_ctime = CURRENT_TIME;
if (current->fsuid != 0) {
saved_cap = current->cap_effective;
cap_clear(current->cap_effective);
}
err = notify_change(dentry, iap);
if (current->fsuid != 0)
current->cap_effective = saved_cap;
if (err)
goto out_nfserr;
if (EX_ISSYNC(fhp->fh_export))
write_inode_now(inode);
}
err = 0;
out:
return err;
out_nfserr:
err = nfserrno(-err);
goto out;
}
/*
* Create a hardlink
* N.B. After this call _both_ ffhp and tfhp need an fh_put
*/
int
nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp,
char *fname, int len, struct svc_fh *tfhp)
{
struct dentry *ddir, *dnew, *dold;
struct inode *dirp, *dest;
int err;
err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_CREATE);
if (err)
goto out;
err = fh_verify(rqstp, tfhp, S_IFREG, MAY_NOP);
if (err)
goto out;
err = nfserr_perm;
if (!len)
goto out;
ddir = ffhp->fh_dentry;
dirp = ddir->d_inode;
dnew = lookup_dentry(fname, dget(ddir), 0);
err = PTR_ERR(dnew);
if (IS_ERR(dnew))
goto out_nfserr;
/*
* Lock the parent before checking for existence
*/
err = fh_lock_parent(ffhp, dnew);
if (err)
goto out_dput;
err = nfserr_exist;
if (dnew->d_inode)
goto out_unlock;
dold = tfhp->fh_dentry;
dest = dold->d_inode;
err = nfserr_acces;
if (nfsd_iscovered(ddir, ffhp->fh_export))
goto out_unlock;
/* FIXME: nxdev for NFSv3 */
if (dirp->i_dev != dest->i_dev)
goto out_unlock;
err = nfserr_perm;
if (IS_IMMUTABLE(dest) /* || IS_APPEND(dest) */ )
goto out_unlock;
if (!dirp->i_op || !dirp->i_op->link)
goto out_unlock;
DQUOT_INIT(dirp);
err = dirp->i_op->link(dold, dirp, dnew);
DQUOT_DROP(dirp);
if (!err) {
if (EX_ISSYNC(ffhp->fh_export)) {
write_inode_now(dirp);
write_inode_now(dest);
}
} else
err = nfserrno(-err);
out_unlock:
fh_unlock(ffhp);
out_dput:
dput(dnew);
out:
return err;
out_nfserr:
err = nfserrno(-err);
goto out;
}
/*
* Create a symlink and look up its inode
* N.B. After this call _both_ fhp and resfhp need an fh_put
*/
int
nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp,
char *fname, int flen,
char *path, int plen,
struct svc_fh *resfhp)
{
struct dentry *dentry, *dnew;
struct inode *dirp;
int err;
err = nfserr_noent;
if (!flen || !plen)
goto out;
err = fh_verify(rqstp, fhp, S_IFDIR, MAY_CREATE);
if (err)
goto out;
dentry = fhp->fh_dentry;
err = nfserr_perm;
if (nfsd_iscovered(dentry, fhp->fh_export))
goto out;
dirp = dentry->d_inode;
if (!dirp->i_op || !dirp->i_op->symlink)
goto out;
dnew = lookup_dentry(fname, dget(dentry), 0);
err = PTR_ERR(dnew);
if (IS_ERR(dnew))
goto out_nfserr;
/*
* Lock the parent before checking for existence
*/
err = fh_lock_parent(fhp, dnew);
if (err)
goto out_compose;
err = nfserr_exist;
if (!dnew->d_inode) {
DQUOT_INIT(dirp);
err = dirp->i_op->symlink(dirp, dnew, path);
DQUOT_DROP(dirp);
if (!err) {
if (EX_ISSYNC(fhp->fh_export))
write_inode_now(dirp);
} else
err = nfserrno(-err);
}
fh_unlock(fhp);
/* Compose the fh so the dentry will be freed ... */
out_compose:
fh_compose(resfhp, fhp->fh_export, dnew);
out:
return err;
out_nfserr:
err = nfserrno(-err);
goto out;
}
/*
* Create a file (regular, directory, device, fifo); UNIX sockets
* not yet implemented.
* If the response fh has been verified, the parent directory should
* already be locked. Note that the parent directory is left locked.
*
* N.B. Every call to nfsd_create needs an fh_put for _both_ fhp and resfhp
*/
int
nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
char *fname, int flen, struct iattr *iap,
int type, dev_t rdev, struct svc_fh *resfhp)
{
struct dentry *dentry, *dchild;
struct inode *dirp;
nfsd_dirop_t opfunc = NULL;
int err;
err = nfserr_perm;
if (!flen)
goto out;
err = fh_verify(rqstp, fhp, S_IFDIR, MAY_CREATE);
if (err)
goto out;
dentry = fhp->fh_dentry;
dirp = dentry->d_inode;
err = nfserr_notdir;
if(!dirp->i_op || !dirp->i_op->lookup)
goto out;
/*
* Check whether the response file handle has been verified yet.
* If it has, the parent directory should already be locked.
*/
if (!resfhp->fh_dverified) {
dchild = lookup_dentry(fname, dget(dentry), 0);
err = PTR_ERR(dchild);
if (IS_ERR(dchild))
goto out_nfserr;
fh_compose(resfhp, fhp->fh_export, dchild);
/* Lock the parent and check for errors ... */
err = fh_lock_parent(fhp, dchild);
if (err)
goto out;
} else {
dchild = resfhp->fh_dentry;
if (!fhp->fh_locked)
printk(KERN_ERR
"nfsd_create: parent %s/%s not locked!\n",
dentry->d_parent->d_name.name,
dentry->d_name.name);
}
/*
* Make sure the child dentry is still negative ...
*/
err = nfserr_exist;
if (dchild->d_inode) {
printk(KERN_WARNING
"nfsd_create: dentry %s/%s not negative!\n",
dentry->d_name.name, dchild->d_name.name);
goto out;
}
/*
* Get the dir op function pointer.
*/
err = nfserr_perm;
switch (type) {
case S_IFREG:
opfunc = (nfsd_dirop_t) dirp->i_op->create;
break;
case S_IFDIR:
opfunc = (nfsd_dirop_t) dirp->i_op->mkdir;
break;
case S_IFCHR:
case S_IFBLK:
/* The client is _NOT_ required to do security enforcement */
if(!capable(CAP_SYS_ADMIN))
{
err = -EPERM;
goto out;
}
case S_IFIFO:
case S_IFSOCK:
opfunc = dirp->i_op->mknod;
break;
}
if (!opfunc)
goto out;
if (!(iap->ia_valid & ATTR_MODE))
iap->ia_mode = 0;
/*
* Call the dir op function to create the object.
*/
DQUOT_INIT(dirp);
err = opfunc(dirp, dchild, iap->ia_mode, rdev);
DQUOT_DROP(dirp);
if (err < 0)
goto out_nfserr;
if (EX_ISSYNC(fhp->fh_export))
write_inode_now(dirp);
/*
* Update the file handle to get the new inode info.
*/
fh_update(resfhp);
/* Set file attributes. Mode has already been set and
* setting uid/gid works only for root. Irix appears to
* send along the gid when it tries to implement setgid
* directories via NFS.
*/
err = 0;
if ((iap->ia_valid &= (ATTR_UID|ATTR_GID|ATTR_MODE)) != 0)
err = nfsd_setattr(rqstp, resfhp, iap);
out:
return err;
out_nfserr:
err = nfserrno(-err);
goto out;
}
/*
* Write data to a file.
* The stable flag requests synchronous writes.
* N.B. After this call fhp needs an fh_put
*/
int
nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset,
char *buf, unsigned long cnt, int stable)
{
struct svc_export *exp;
struct file file;
struct dentry *dentry;
struct inode *inode;
mm_segment_t oldfs;
int err = 0;
#ifdef CONFIG_QUOTA
uid_t saved_euid;
#endif
if (!cnt)
goto out;
err = nfsd_open(rqstp, fhp, S_IFREG, OPEN_WRITE, &file);
if (err)
goto out;
err = nfserr_perm;
if (!file.f_op->write)
goto out_close;
dentry = file.f_dentry;
inode = dentry->d_inode;
exp = fhp->fh_export;
/*
* Request sync writes if
* - the sync export option has been set, or
* - the client requested O_SYNC behavior (NFSv3 feature).
* When gathered writes have been configured for this volume,
* flushing the data to disk is handled separately below.
*/
if ((stable || (stable = EX_ISSYNC(exp))) && !EX_WGATHER(exp))
file.f_flags |= O_SYNC;
fh_lock(fhp); /* lock inode */
file.f_pos = offset; /* set write offset */
/* Write the data. */
oldfs = get_fs(); set_fs(KERNEL_DS);
#ifdef CONFIG_QUOTA
/* This is for disk quota. */
saved_euid = current->euid;
current->euid = current->fsuid;
err = file.f_op->write(&file, buf, cnt, &file.f_pos);
current->euid = saved_euid;
#else
err = file.f_op->write(&file, buf, cnt, &file.f_pos);
#endif
set_fs(oldfs);
/* clear setuid/setgid flag after write */
if (err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID))) {
struct iattr ia;
kernel_cap_t saved_cap;
ia.ia_valid = ATTR_MODE;
ia.ia_mode = inode->i_mode & ~(S_ISUID | S_ISGID);
if (current->fsuid != 0) {
saved_cap = current->cap_effective;
cap_clear(current->cap_effective);
}
notify_change(dentry, &ia);
if (current->fsuid != 0)
current->cap_effective = saved_cap;
}
fh_unlock(fhp); /* unlock inode */
if (err >= 0 && stable) {
static unsigned long last_ino = 0;
static kdev_t last_dev = NODEV;
/*
* Gathered writes: If another process is currently
* writing to the file, there's a high chance
* this is another nfsd (triggered by a bulk write
* from a client's biod). Rather than syncing the
* file with each write request, we sleep for 10 msec.
*
* I don't know if this roughly approximates
* C. Juszak's idea of gathered writes, but it's a
* nice and simple solution (IMHO), and it seems to
* work:-)
*/
if (EX_WGATHER(exp) && (inode->i_writecount > 1
|| (last_ino == inode->i_ino && last_dev == inode->i_dev))) {
#if 0
interruptible_sleep_on_timeout(&inode->i_wait, 10 * HZ / 1000);
#else
dprintk("nfsd: write defer %d\n", current->pid);
schedule_timeout((HZ+99)/100);
dprintk("nfsd: write resume %d\n", current->pid);
#endif
}
if (inode->i_state & I_DIRTY) {
dprintk("nfsd: write sync %d\n", current->pid);
nfsd_sync(inode, &file);
write_inode_now(inode);
}
wake_up(&inode->i_wait);
last_ino = inode->i_ino;
last_dev = inode->i_dev;
}
dprintk("nfsd: write complete\n");
if (err >= 0)
err = 0;
else
err = nfserrno(-err);
out_close:
nfsd_close(&file);
out:
return err;
}
