int notify_change(struct dentry * dentry, struct iattr * attr)
{
struct inode *inode = dentry->d_inode;
int error;
struct timespec now;
unsigned int ia_valid = attr->ia_valid;
mode_t ia_mode = attr->ia_mode;
now = current_fs_time(inode->i_sb);
attr->ia_ctime = now;
if (!(ia_valid & ATTR_ATIME_SET))
attr->ia_atime = now;
if (!(ia_valid & ATTR_MTIME_SET))
attr->ia_mtime = now;
if (ia_valid & ATTR_KILL_SUID) {
ia_valid &= ~ATTR_KILL_SUID;
if (inode->i_mode & S_ISUID) {
if (!(ia_valid & ATTR_MODE)) {
ia_valid |= ATTR_MODE;
ia_mode = inode->i_mode;
}
ia_mode &= ~S_ISUID;
}
}
if (ia_valid & ATTR_KILL_SGID) {
ia_valid &= ~ATTR_KILL_SGID;
if ((inode->i_mode & (S_ISGID | S_IXGRP)) ==
(S_ISGID | S_IXGRP)) {
if (!(ia_valid & ATTR_MODE)) {
ia_valid |= ATTR_MODE;
ia_mode = inode->i_mode;
}
ia_mode &= ~S_ISGID;
}
}
if (!ia_valid)
return 0;
/*
* For RHEL, we've added the S_NOATTRKILL flag to allow filesystems
* to opt-out of ATTR_KILL_S*ID processing. The ATTR_KILL_S*ID bits
* are now handled in two stages. First, we calculate what the
* ia_valid and the ia_mode would look like if we were to allow the
* ATTR_KILL_S*ID bits to modify them. We then make the decision of
* whether to allow the modification to occur. We could just skip
* all of the ATTR_KILL_S*ID processing altogether, but we need it
* for inotify. If a process is watching for mode changes, we want
* it to be notified if we suspect that the server will be doing the
* mode change for us.
*/
if ((ia_valid & ATTR_MODE) && !(inode->i_flags & S_NOATTRKILL)) {
attr->ia_valid = ia_valid;
attr->ia_mode = ia_mode;
}
if (ia_valid & ATTR_SIZE)
down_write(&dentry->d_inode->i_alloc_sem);
if (inode->i_op && inode->i_op->setattr) {
error = security_inode_setattr(dentry, attr);
if (!error)
error = inode->i_op->setattr(dentry, attr);
} else {
error = inode_change_ok(inode, attr);
if (!error)
error = security_inode_setattr(dentry, attr);
if (!error) {
if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) ||
(ia_valid & ATTR_GID && attr->ia_gid != inode->i_gid))
error = DQUOT_TRANSFER(inode, attr) ? -EDQUOT : 0;
if (!error)
error = inode_setattr(inode, attr);
}
}
if (ia_valid & ATTR_SIZE)
up_write(&dentry->d_inode->i_alloc_sem);
if (!error)
fsnotify_change(dentry, ia_valid);
return error;
}
dotraplinkage void __kprobes
do_general_protection(struct pt_regs *regs, long error_code)
{
struct task_struct *tsk;
conditional_sti(regs);
#ifdef CONFIG_X86_32
if (v8086_mode(regs))
goto gp_in_vm86;
#endif
tsk = current;
if (!user_mode_novm(regs))
goto gp_in_kernel;
#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC)
if (!nx_enabled && tsk->mm && (tsk->mm->pax_flags & MF_PAX_PAGEEXEC)) {
struct mm_struct *mm = tsk->mm;
unsigned long limit;
down_write(&mm->mmap_sem);
limit = mm->context.user_cs_limit;
if (limit < TASK_SIZE) {
track_exec_limit(mm, limit, TASK_SIZE, VM_EXEC);
up_write(&mm->mmap_sem);
return;
}
up_write(&mm->mmap_sem);
}
#endif
tsk->thread.error_code = error_code;
tsk->thread.trap_no = 13;
if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
printk_ratelimit()) {
printk(KERN_INFO
"%s[%d] general protection ip:%lx sp:%lx error:%lx",
tsk->comm, task_pid_nr(tsk),
regs->ip, regs->sp, error_code);
print_vma_addr(" in ", regs->ip);
printk("\n");
}
force_sig(SIGSEGV, tsk);
return;
#ifdef CONFIG_X86_32
gp_in_vm86:
local_irq_enable();
handle_vm86_fault((struct kernel_vm86_regs *) regs, error_code);
return;
#endif
gp_in_kernel:
if (fixup_exception(regs))
return;
tsk->thread.error_code = error_code;
tsk->thread.trap_no = 13;
if (notify_die(DIE_GPF, "general protection fault", regs,
error_code, 13, SIGSEGV) == NOTIFY_STOP)
return;
#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC)
if ((regs->cs & 0xFFFF) == __KERNEL_CS)
die("PAX: suspicious general protection fault", regs, error_code);
else
#endif
die("general protection fault", regs, error_code);
}
static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm)
{
struct vm_area_struct *mpnt, *tmp, *prev, **pprev;
struct rb_node **rb_link, *rb_parent;
int retval;
unsigned long charge;
struct mempolicy *pol;
uprobe_start_dup_mmap();
down_write(&oldmm->mmap_sem);
flush_cache_dup_mm(oldmm);
uprobe_dup_mmap(oldmm, mm);
/*
* Not linked in yet - no deadlock potential:
*/
down_write_nested(&mm->mmap_sem, SINGLE_DEPTH_NESTING);
mm->locked_vm = 0;
mm->mmap = NULL;
mm->mmap_cache = NULL;
mm->free_area_cache = oldmm->mmap_base;
mm->cached_hole_size = ~0UL;
mm->map_count = 0;
cpumask_clear(mm_cpumask(mm));
mm->mm_rb = RB_ROOT;
rb_link = &mm->mm_rb.rb_node;
rb_parent = NULL;
pprev = &mm->mmap;
retval = ksm_fork(mm, oldmm);
if (retval)
goto out;
retval = khugepaged_fork(mm, oldmm);
if (retval)
goto out;
prev = NULL;
for (mpnt = oldmm->mmap; mpnt; mpnt = mpnt->vm_next) {
struct file *file;
if (mpnt->vm_flags & VM_DONTCOPY) {
vm_stat_account(mm, mpnt->vm_flags, mpnt->vm_file,
-vma_pages(mpnt));
continue;
}
charge = 0;
if (mpnt->vm_flags & VM_ACCOUNT) {
unsigned long len = vma_pages(mpnt);
if (security_vm_enough_memory_mm(oldmm, len)) /* sic */
goto fail_nomem;
charge = len;
}
tmp = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL);
if (!tmp)
goto fail_nomem;
*tmp = *mpnt;
INIT_LIST_HEAD(&tmp->anon_vma_chain);
pol = mpol_dup(vma_policy(mpnt));
retval = PTR_ERR(pol);
if (IS_ERR(pol))
goto fail_nomem_policy;
vma_set_policy(tmp, pol);
tmp->vm_mm = mm;
if (anon_vma_fork(tmp, mpnt))
goto fail_nomem_anon_vma_fork;
tmp->vm_flags &= ~VM_LOCKED;
tmp->vm_next = tmp->vm_prev = NULL;
file = tmp->vm_file;
if (file) {
struct inode *inode = file->f_path.dentry->d_inode;
struct address_space *mapping = file->f_mapping;
get_file(file);
if (tmp->vm_flags & VM_DENYWRITE)
atomic_dec(&inode->i_writecount);
mutex_lock(&mapping->i_mmap_mutex);
if (tmp->vm_flags & VM_SHARED)
mapping->i_mmap_writable++;
flush_dcache_mmap_lock(mapping);
/* insert tmp into the share list, just after mpnt */
if (unlikely(tmp->vm_flags & VM_NONLINEAR))
vma_nonlinear_insert(tmp,
&mapping->i_mmap_nonlinear);
else
vma_interval_tree_insert_after(tmp, mpnt,
&mapping->i_mmap);
flush_dcache_mmap_unlock(mapping);
mutex_unlock(&mapping->i_mmap_mutex);
}
/*
* Clear hugetlb-related page reserves for children. This only
* affects MAP_PRIVATE mappings. Faults generated by the child
* are not guaranteed to succeed, even if read-only
*/
if (is_vm_hugetlb_page(tmp))
reset_vma_resv_huge_pages(tmp);
/*
* Link in the new vma and copy the page table entries.
*/
*pprev = tmp;
pprev = &tmp->vm_next;
tmp->vm_prev = prev;
prev = tmp;
__vma_link_rb(mm, tmp, rb_link, rb_parent);
rb_link = &tmp->vm_rb.rb_right;
rb_parent = &tmp->vm_rb;
mm->map_count++;
retval = copy_page_range(mm, oldmm, mpnt);
if (tmp->vm_ops && tmp->vm_ops->open)
tmp->vm_ops->open(tmp);
if (retval)
goto out;
}
/* a new mm has just been created */
arch_dup_mmap(oldmm, mm);
retval = 0;
out:
up_write(&mm->mmap_sem);
flush_tlb_mm(oldmm);
up_write(&oldmm->mmap_sem);
uprobe_end_dup_mmap();
return retval;
fail_nomem_anon_vma_fork:
mpol_put(pol);
fail_nomem_policy:
kmem_cache_free(vm_area_cachep, tmp);
fail_nomem:
retval = -ENOMEM;
vm_unacct_memory(charge);
goto out;
}
static void gameport_release_driver(struct gameport *gameport)
{
down_write(&gameport_bus.subsys.rwsem);
device_release_driver(&gameport->dev);
up_write(&gameport_bus.subsys.rwsem);
}
void
ia64_elf32_init (struct pt_regs *regs)
{
struct vm_area_struct *vma;
/*
* Map GDT and TSS below 4GB, where the processor can find them. We need to map
* it with privilege level 3 because the IVE uses non-privileged accesses to these
* tables. IA-32 segmentation is used to protect against IA-32 accesses to them.
*/
vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL);
if (vma) {
vma->vm_mm = current->mm;
vma->vm_start = IA32_GDT_OFFSET;
vma->vm_end = vma->vm_start + max(PAGE_SIZE, 2*IA32_PAGE_SIZE);
vma->vm_page_prot = PAGE_SHARED;
vma->vm_flags = VM_READ|VM_MAYREAD;
vma->vm_ops = &ia32_shared_page_vm_ops;
vma->vm_pgoff = 0;
vma->vm_file = NULL;
vma->vm_private_data = NULL;
down_write(¤t->mm->mmap_sem);
{
insert_vm_struct(current->mm, vma);
}
up_write(¤t->mm->mmap_sem);
}
/*
* Install LDT as anonymous memory. This gives us all-zero segment descriptors
* until a task modifies them via modify_ldt().
*/
vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL);
if (vma) {
vma->vm_mm = current->mm;
vma->vm_start = IA32_LDT_OFFSET;
vma->vm_end = vma->vm_start + PAGE_ALIGN(IA32_LDT_ENTRIES*IA32_LDT_ENTRY_SIZE);
vma->vm_page_prot = PAGE_SHARED;
vma->vm_flags = VM_READ|VM_WRITE|VM_MAYREAD|VM_MAYWRITE;
vma->vm_ops = NULL;
vma->vm_pgoff = 0;
vma->vm_file = NULL;
vma->vm_private_data = NULL;
down_write(¤t->mm->mmap_sem);
{
insert_vm_struct(current->mm, vma);
}
up_write(¤t->mm->mmap_sem);
}
ia64_psr(regs)->ac = 0; /* turn off alignment checking */
regs->loadrs = 0;
/*
* According to the ABI %edx points to an `atexit' handler. Since we don't have
* one we'll set it to 0 and initialize all the other registers just to make
* things more deterministic, ala the i386 implementation.
*/
regs->r8 = 0; /* %eax */
regs->r11 = 0; /* %ebx */
regs->r9 = 0; /* %ecx */
regs->r10 = 0; /* %edx */
regs->r13 = 0; /* %ebp */
regs->r14 = 0; /* %esi */
regs->r15 = 0; /* %edi */
current->thread.eflag = IA32_EFLAG;
current->thread.fsr = IA32_FSR_DEFAULT;
current->thread.fcr = IA32_FCR_DEFAULT;
current->thread.fir = 0;
current->thread.fdr = 0;
/*
* Setup GDTD. Note: GDTD is the descrambled version of the pseudo-descriptor
* format defined by Figure 3-11 "Pseudo-Descriptor Format" in the IA-32
* architecture manual. Also note that the only fields that are not ignored are
* `base', `limit', 'G', `P' (must be 1) and `S' (must be 0).
*/
regs->r31 = IA32_SEG_UNSCRAMBLE(IA32_SEG_DESCRIPTOR(IA32_GDT_OFFSET, IA32_PAGE_SIZE - 1,
0, 0, 0, 1, 0, 0, 0));
/* Setup the segment selectors */
regs->r16 = (__USER_DS << 16) | __USER_DS; /* ES == DS, GS, FS are zero */
regs->r17 = (__USER_DS << 16) | __USER_CS; /* SS, CS; ia32_load_state() sets TSS and LDT */
ia32_load_segment_descriptors(current);
ia32_load_state(current);
}
/**
* nilfs_resize_fs - resize the filesystem
* @sb: super block instance
* @newsize: new size of the filesystem (in bytes)
*/
int nilfs_resize_fs(struct super_block *sb, __u64 newsize)
{
struct the_nilfs *nilfs = sb->s_fs_info;
struct nilfs_super_block **sbp;
__u64 devsize, newnsegs;
loff_t sb2off;
int ret;
ret = -ERANGE;
devsize = i_size_read(sb->s_bdev->bd_inode);
if (newsize > devsize)
goto out;
/*
* Write lock is required to protect some functions depending
* on the number of segments, the number of reserved segments,
* and so forth.
*/
down_write(&nilfs->ns_segctor_sem);
sb2off = NILFS_SB2_OFFSET_BYTES(newsize);
newnsegs = sb2off >> nilfs->ns_blocksize_bits;
do_div(newnsegs, nilfs->ns_blocks_per_segment);
ret = nilfs_sufile_resize(nilfs->ns_sufile, newnsegs);
up_write(&nilfs->ns_segctor_sem);
if (ret < 0)
goto out;
ret = nilfs_construct_segment(sb);
if (ret < 0)
goto out;
down_write(&nilfs->ns_sem);
nilfs_move_2nd_super(sb, sb2off);
ret = -EIO;
sbp = nilfs_prepare_super(sb, 0);
if (likely(sbp)) {
nilfs_set_log_cursor(sbp[0], nilfs);
/*
* Drop NILFS_RESIZE_FS flag for compatibility with
* mount-time resize which may be implemented in a
* future release.
*/
sbp[0]->s_state = cpu_to_le16(le16_to_cpu(sbp[0]->s_state) &
~NILFS_RESIZE_FS);
sbp[0]->s_dev_size = cpu_to_le64(newsize);
sbp[0]->s_nsegments = cpu_to_le64(nilfs->ns_nsegments);
if (sbp[1])
memcpy(sbp[1], sbp[0], nilfs->ns_sbsize);
ret = nilfs_commit_super(sb, NILFS_SB_COMMIT_ALL);
}
up_write(&nilfs->ns_sem);
/*
* Reset the range of allocatable segments last. This order
* is important in the case of expansion because the secondary
* superblock must be protected from log write until migration
* completes.
*/
if (!ret)
nilfs_sufile_set_alloc_range(nilfs->ns_sufile, 0, newnsegs - 1);
out:
return ret;
}
void ext4_inline_data_truncate(struct inode *inode, int *has_inline)
{
handle_t *handle;
int inline_size, value_len, needed_blocks;
size_t i_size;
void *value = NULL;
struct ext4_xattr_ibody_find is = {
.s = { .not_found = -ENODATA, },
};
struct ext4_xattr_info i = {
.name_index = EXT4_XATTR_INDEX_SYSTEM,
.name = EXT4_XATTR_SYSTEM_DATA,
};
needed_blocks = ext4_writepage_trans_blocks(inode);
handle = ext4_journal_start(inode, EXT4_HT_INODE, needed_blocks);
if (IS_ERR(handle))
return;
down_write(&EXT4_I(inode)->xattr_sem);
if (!ext4_has_inline_data(inode)) {
*has_inline = 0;
ext4_journal_stop(handle);
return;
}
if (ext4_orphan_add(handle, inode))
goto out;
if (ext4_get_inode_loc(inode, &is.iloc))
goto out;
down_write(&EXT4_I(inode)->i_data_sem);
i_size = inode->i_size;
inline_size = ext4_get_inline_size(inode);
EXT4_I(inode)->i_disksize = i_size;
if (i_size < inline_size) {
/* Clear the content in the xattr space. */
if (inline_size > EXT4_MIN_INLINE_DATA_SIZE) {
if (ext4_xattr_ibody_find(inode, &i, &is))
goto out_error;
BUG_ON(is.s.not_found);
value_len = le32_to_cpu(is.s.here->e_value_size);
value = kmalloc(value_len, GFP_NOFS);
if (!value)
goto out_error;
if (ext4_xattr_ibody_get(inode, i.name_index, i.name,
value, value_len))
goto out_error;
i.value = value;
i.value_len = i_size > EXT4_MIN_INLINE_DATA_SIZE ?
i_size - EXT4_MIN_INLINE_DATA_SIZE : 0;
if (ext4_xattr_ibody_inline_set(handle, inode, &i, &is))
goto out_error;
}
/* Clear the content within i_blocks. */
if (i_size < EXT4_MIN_INLINE_DATA_SIZE) {
void *p = (void *) ext4_raw_inode(&is.iloc)->i_block;
memset(p + i_size, 0,
EXT4_MIN_INLINE_DATA_SIZE - i_size);
}
EXT4_I(inode)->i_inline_size = i_size <
EXT4_MIN_INLINE_DATA_SIZE ?
EXT4_MIN_INLINE_DATA_SIZE : i_size;
}
out_error:
up_write(&EXT4_I(inode)->i_data_sem);
out:
brelse(is.iloc.bh);
up_write(&EXT4_I(inode)->xattr_sem);
kfree(value);
if (inode->i_nlink)
ext4_orphan_del(handle, inode);
inode->i_mtime = inode->i_ctime = ext4_current_time(inode);
ext4_mark_inode_dirty(handle, inode);
if (IS_SYNC(inode))
ext4_handle_sync(handle);
ext4_journal_stop(handle);
return;
}
static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs)
{
struct exec ex;
unsigned long error;
unsigned long fd_offset;
unsigned long rlim;
int retval;
ex = *((struct exec *) bprm->buf); /* exec-header */
if ((N_MAGIC(ex) != ZMAGIC && N_MAGIC(ex) != OMAGIC &&
N_MAGIC(ex) != QMAGIC && N_MAGIC(ex) != NMAGIC) ||
N_TRSIZE(ex) || N_DRSIZE(ex) ||
bprm->file->f_dentry->d_inode->i_size < ex.a_text+ex.a_data+N_SYMSIZE(ex)+N_TXTOFF(ex)) {
return -ENOEXEC;
}
fd_offset = N_TXTOFF(ex);
/* Check initial limits. This avoids letting people circumvent
* size limits imposed on them by creating programs with large
* arrays in the data or bss.
*/
rlim = current->rlim[RLIMIT_DATA].rlim_cur;
if (rlim >= RLIM_INFINITY)
rlim = ~0;
if (ex.a_data + ex.a_bss > rlim)
return -ENOMEM;
/* Flush all traces of the currently running executable */
retval = flush_old_exec(bprm);
if (retval)
return retval;
/* OK, This is the point of no return */
#if defined(__alpha__)
SET_AOUT_PERSONALITY(bprm, ex);
#elif defined(__sparc__)
set_personality(PER_SUNOS);
#if !defined(__sparc_v9__)
memcpy(¤t->thread.core_exec, &ex, sizeof(struct exec));
#endif
#else
set_personality(PER_LINUX);
#endif
current->mm->end_code = ex.a_text +
(current->mm->start_code = N_TXTADDR(ex));
current->mm->end_data = ex.a_data +
(current->mm->start_data = N_DATADDR(ex));
current->mm->brk = ex.a_bss +
(current->mm->start_brk = N_BSSADDR(ex));
current->mm->rss = 0;
current->mm->mmap = NULL;
#ifdef CONFIG_ARM_FASS
arch_new_mm(current, current->mm);
#endif
compute_creds(bprm);
current->flags &= ~PF_FORKNOEXEC;
#ifdef __sparc__
if (N_MAGIC(ex) == NMAGIC) {
loff_t pos = fd_offset;
/* F**k me plenty... */
/* <AOL></AOL> */
down_write(¤t->mm->mmap_sem);
error = do_brk(N_TXTADDR(ex), ex.a_text);
up_write(¤t->mm->mmap_sem);
bprm->file->f_op->read(bprm->file, (char *) N_TXTADDR(ex),
ex.a_text, &pos);
down_write(¤t->mm->mmap_sem);
error = do_brk(N_DATADDR(ex), ex.a_data);
up_write(¤t->mm->mmap_sem);
bprm->file->f_op->read(bprm->file, (char *) N_DATADDR(ex),
ex.a_data, &pos);
goto beyond_if;
}
#endif
if (N_MAGIC(ex) == OMAGIC) {
unsigned long text_addr, map_size;
loff_t pos;
text_addr = N_TXTADDR(ex);
#if defined(__alpha__) || defined(__sparc__)
pos = fd_offset;
map_size = ex.a_text+ex.a_data + PAGE_SIZE - 1;
#else
pos = 32;
map_size = ex.a_text+ex.a_data;
#endif
down_write(¤t->mm->mmap_sem);
error = do_brk(text_addr & PAGE_MASK, map_size);
up_write(¤t->mm->mmap_sem);
if (error != (text_addr & PAGE_MASK)) {
send_sig(SIGKILL, current, 0);
return error;
}
error = bprm->file->f_op->read(bprm->file, (char *)text_addr,
ex.a_text+ex.a_data, &pos);
if ((signed long)error < 0) {
send_sig(SIGKILL, current, 0);
return error;
}
flush_icache_range(text_addr, text_addr+ex.a_text+ex.a_data);
} else {
static unsigned long error_time, error_time2;
if ((ex.a_text & 0xfff || ex.a_data & 0xfff) &&
(N_MAGIC(ex) != NMAGIC) && (jiffies-error_time2) > 5*HZ)
{
printk(KERN_NOTICE "executable not page aligned\n");
error_time2 = jiffies;
}
if ((fd_offset & ~PAGE_MASK) != 0 &&
(jiffies-error_time) > 5*HZ)
{
printk(KERN_WARNING
"fd_offset is not page aligned. Please convert program: %s\n",
bprm->file->f_dentry->d_name.name);
error_time = jiffies;
}
if (!bprm->file->f_op->mmap||((fd_offset & ~PAGE_MASK) != 0)) {
loff_t pos = fd_offset;
down_write(¤t->mm->mmap_sem);
do_brk(N_TXTADDR(ex), ex.a_text+ex.a_data);
up_write(¤t->mm->mmap_sem);
bprm->file->f_op->read(bprm->file,(char *)N_TXTADDR(ex),
ex.a_text+ex.a_data, &pos);
flush_icache_range((unsigned long) N_TXTADDR(ex),
(unsigned long) N_TXTADDR(ex) +
ex.a_text+ex.a_data);
goto beyond_if;
}
down_write(¤t->mm->mmap_sem);
error = do_mmap(bprm->file, N_TXTADDR(ex), ex.a_text,
PROT_READ | PROT_EXEC,
MAP_FIXED | MAP_PRIVATE | MAP_DENYWRITE | MAP_EXECUTABLE,
fd_offset);
up_write(¤t->mm->mmap_sem);
if (error != N_TXTADDR(ex)) {
send_sig(SIGKILL, current, 0);
return error;
}
down_write(¤t->mm->mmap_sem);
error = do_mmap(bprm->file, N_DATADDR(ex), ex.a_data,
PROT_READ | PROT_WRITE | PROT_EXEC,
MAP_FIXED | MAP_PRIVATE | MAP_DENYWRITE | MAP_EXECUTABLE,
fd_offset + ex.a_text);
up_write(¤t->mm->mmap_sem);
if (error != N_DATADDR(ex)) {
send_sig(SIGKILL, current, 0);
return error;
}
}
beyond_if:
set_binfmt(&aout_format);
retval = set_brk(current->mm->start_brk, current->mm->brk);
if (retval < 0) {
send_sig(SIGKILL, current, 0);
return retval;
}
retval = setup_arg_pages(bprm, STACK_TOP);
if (retval < 0) {
/* Someone check-me: is this error path enough? */
send_sig(SIGKILL, current, 0);
return retval;
}
current->mm->start_stack =
(unsigned long) create_aout_tables((char *) bprm->p, bprm);
#ifdef __alpha__
regs->gp = ex.a_gpvalue;
#endif
start_thread(regs, ex.a_entry, current->mm->start_stack);
if (current->ptrace & PT_PTRACED)
send_sig(SIGTRAP, current, 0);
#ifndef __arm__
return 0;
#else
return regs->ARM_r0;
#endif
}
static int load_aout_library(struct file *file)
{
struct inode * inode;
unsigned long bss, start_addr, len;
unsigned long error;
int retval;
struct exec ex;
inode = file->f_dentry->d_inode;
retval = -ENOEXEC;
error = kernel_read(file, 0, (char *) &ex, sizeof(ex));
if (error != sizeof(ex))
goto out;
/* We come in here for the regular a.out style of shared libraries */
if ((N_MAGIC(ex) != ZMAGIC && N_MAGIC(ex) != QMAGIC) || N_TRSIZE(ex) ||
N_DRSIZE(ex) || ((ex.a_entry & 0xfff) && N_MAGIC(ex) == ZMAGIC) ||
inode->i_size < ex.a_text+ex.a_data+N_SYMSIZE(ex)+N_TXTOFF(ex)) {
goto out;
}
if (N_FLAGS(ex))
goto out;
/* For QMAGIC, the starting address is 0x20 into the page. We mask
this off to get the starting address for the page */
#ifndef __arm__
start_addr = ex.a_entry & 0xfffff000;
#else
start_addr = ex.a_entry & 0xffff8000;
#endif
if ((N_TXTOFF(ex) & ~PAGE_MASK) != 0) {
static unsigned long error_time;
loff_t pos = N_TXTOFF(ex);
if ((jiffies-error_time) > 5*HZ)
{
printk(KERN_WARNING
"N_TXTOFF is not page aligned. Please convert library: %s\n",
file->f_dentry->d_name.name);
error_time = jiffies;
}
down_write(¤t->mm->mmap_sem);
do_brk(start_addr, ex.a_text + ex.a_data + ex.a_bss);
up_write(¤t->mm->mmap_sem);
file->f_op->read(file, (char *)start_addr,
ex.a_text + ex.a_data, &pos);
flush_icache_range((unsigned long) start_addr,
(unsigned long) start_addr + ex.a_text + ex.a_data);
retval = 0;
goto out;
}
/* Now use mmap to map the library into memory. */
down_write(¤t->mm->mmap_sem);
error = do_mmap(file, start_addr, ex.a_text + ex.a_data,
PROT_READ | PROT_WRITE | PROT_EXEC,
MAP_FIXED | MAP_PRIVATE | MAP_DENYWRITE,
N_TXTOFF(ex));
up_write(¤t->mm->mmap_sem);
retval = error;
if (error != start_addr)
goto out;
len = PAGE_ALIGN(ex.a_text + ex.a_data);
bss = ex.a_text + ex.a_data + ex.a_bss;
if (bss > len) {
down_write(¤t->mm->mmap_sem);
error = do_brk(start_addr + len, bss - len);
up_write(¤t->mm->mmap_sem);
retval = error;
if (error != start_addr + len)
goto out;
}
retval = 0;
out:
return retval;
}
/*
* We need to be able to change a mapping table under a mounted
* filesystem. For example we might want to move some data in
* the background. Before the table can be swapped with
* dm_bind_table, dm_suspend must be called to flush any in
* flight bios and ensure that any further io gets deferred.
*/
int dm_suspend(struct mapped_device *md, int do_lockfs)
{
struct dm_table *map = NULL;
DECLARE_WAITQUEUE(wait, current);
struct bio *def;
int r = -EINVAL;
down(&md->suspend_lock);
if (dm_suspended(md))
goto out;
map = dm_get_table(md);
/* This does not get reverted if there's an error later. */
dm_table_presuspend_targets(map);
md->suspended_bdev = bdget_disk(md->disk, 0);
if (!md->suspended_bdev) {
DMWARN("bdget failed in dm_suspend");
r = -ENOMEM;
goto out;
}
/* Flush I/O to the device. */
if (do_lockfs) {
r = lock_fs(md);
if (r)
goto out;
}
/*
* First we set the BLOCK_IO flag so no more ios will be mapped.
*/
down_write(&md->io_lock);
set_bit(DMF_BLOCK_IO, &md->flags);
add_wait_queue(&md->wait, &wait);
up_write(&md->io_lock);
/* unplug */
if (map)
dm_table_unplug_all(map);
/*
* Then we wait for the already mapped ios to
* complete.
*/
while (1) {
set_current_state(TASK_INTERRUPTIBLE);
if (!atomic_read(&md->pending) || signal_pending(current))
break;
io_schedule();
}
set_current_state(TASK_RUNNING);
down_write(&md->io_lock);
remove_wait_queue(&md->wait, &wait);
/* were we interrupted ? */
r = -EINTR;
if (atomic_read(&md->pending)) {
clear_bit(DMF_BLOCK_IO, &md->flags);
def = bio_list_get(&md->deferred);
__flush_deferred_io(md, def);
up_write(&md->io_lock);
unlock_fs(md);
goto out;
}
up_write(&md->io_lock);
dm_table_postsuspend_targets(map);
set_bit(DMF_SUSPENDED, &md->flags);
r = 0;
out:
if (r && md->suspended_bdev) {
bdput(md->suspended_bdev);
md->suspended_bdev = NULL;
}
dm_table_put(map);
up(&md->suspend_lock);
return r;
}
/*
* inode->i_mutex: down
*/
int
reiserfs_xattr_set_handle(struct reiserfs_transaction_handle *th,
struct inode *inode, const char *name,
const void *buffer, size_t buffer_size, int flags)
{
int err = 0;
struct dentry *dentry;
struct page *page;
char *data;
size_t file_pos = 0;
size_t buffer_pos = 0;
size_t new_size;
__u32 xahash = 0;
if (get_inode_sd_version(inode) == STAT_DATA_V1)
return -EOPNOTSUPP;
if (!buffer) {
err = lookup_and_delete_xattr(inode, name);
return err;
}
dentry = xattr_lookup(inode, name, flags);
if (IS_ERR(dentry))
return PTR_ERR(dentry);
down_write(&REISERFS_I(inode)->i_xattr_sem);
xahash = xattr_hash(buffer, buffer_size);
while (buffer_pos < buffer_size || buffer_pos == 0) {
size_t chunk;
size_t skip = 0;
size_t page_offset = (file_pos & (PAGE_CACHE_SIZE - 1));
if (buffer_size - buffer_pos > PAGE_CACHE_SIZE)
chunk = PAGE_CACHE_SIZE;
else
chunk = buffer_size - buffer_pos;
page = reiserfs_get_page(dentry->d_inode, file_pos);
if (IS_ERR(page)) {
err = PTR_ERR(page);
goto out_unlock;
}
lock_page(page);
data = page_address(page);
if (file_pos == 0) {
struct reiserfs_xattr_header *rxh;
skip = file_pos = sizeof(struct reiserfs_xattr_header);
if (chunk + skip > PAGE_CACHE_SIZE)
chunk = PAGE_CACHE_SIZE - skip;
rxh = (struct reiserfs_xattr_header *)data;
rxh->h_magic = cpu_to_le32(REISERFS_XATTR_MAGIC);
rxh->h_hash = cpu_to_le32(xahash);
}
reiserfs_write_lock(inode->i_sb);
err = __reiserfs_write_begin(page, page_offset, chunk + skip);
if (!err) {
if (buffer)
memcpy(data + skip, buffer + buffer_pos, chunk);
err = reiserfs_commit_write(NULL, page, page_offset,
page_offset + chunk +
skip);
}
reiserfs_write_unlock(inode->i_sb);
unlock_page(page);
reiserfs_put_page(page);
buffer_pos += chunk;
file_pos += chunk;
skip = 0;
if (err || buffer_size == 0 || !buffer)
break;
}
new_size = buffer_size + sizeof(struct reiserfs_xattr_header);
if (!err && new_size < i_size_read(dentry->d_inode)) {
struct iattr newattrs = {
.ia_ctime = current_fs_time(inode->i_sb),
.ia_size = new_size,
.ia_valid = ATTR_SIZE | ATTR_CTIME,
};
mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_XATTR);
inode_dio_wait(dentry->d_inode);
err = reiserfs_setattr(dentry, &newattrs);
mutex_unlock(&dentry->d_inode->i_mutex);
} else
update_ctime(inode);
out_unlock:
up_write(&REISERFS_I(inode)->i_xattr_sem);
dput(dentry);
return err;
}
static void genl_unlock_all(void)
{
genl_unlock();
up_write(&cb_lock);
}
static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode,
struct inode *tmp_inode)
{
int retval;
__le32 i_data[3];
struct ext4_inode_info *ei = EXT4_I(inode);
struct ext4_inode_info *tmp_ei = EXT4_I(tmp_inode);
/*
* One credit accounted for writing the
* i_data field of the original inode
*/
retval = ext4_journal_extend(handle, 1);
if (retval) {
retval = ext4_journal_restart(handle, 1);
if (retval)
goto err_out;
}
i_data[0] = ei->i_data[EXT4_IND_BLOCK];
i_data[1] = ei->i_data[EXT4_DIND_BLOCK];
i_data[2] = ei->i_data[EXT4_TIND_BLOCK];
down_write(&EXT4_I(inode)->i_data_sem);
/*
* if EXT4_EXT_MIGRATE is cleared a block allocation
* happened after we started the migrate. We need to
* fail the migrate
*/
if (!(EXT4_I(inode)->i_flags & EXT4_EXT_MIGRATE)) {
retval = -EAGAIN;
up_write(&EXT4_I(inode)->i_data_sem);
goto err_out;
} else
EXT4_I(inode)->i_flags = EXT4_I(inode)->i_flags &
~EXT4_EXT_MIGRATE;
/*
* We have the extent map build with the tmp inode.
* Now copy the i_data across
*/
ei->i_flags |= EXT4_EXTENTS_FL;
memcpy(ei->i_data, tmp_ei->i_data, sizeof(ei->i_data));
/*
* Update i_blocks with the new blocks that got
* allocated while adding extents for extent index
* blocks.
*
* While converting to extents we need not
* update the orignal inode i_blocks for extent blocks
* via quota APIs. The quota update happened via tmp_inode already.
*/
spin_lock(&inode->i_lock);
inode->i_blocks += tmp_inode->i_blocks;
spin_unlock(&inode->i_lock);
up_write(&EXT4_I(inode)->i_data_sem);
/*
* We mark the inode dirty after, because we decrement the
* i_blocks when freeing the indirect meta-data blocks
*/
retval = free_ind_block(handle, inode, i_data);
ext4_mark_inode_dirty(handle, inode);
err_out:
return retval;
}
/*
* Allocate a new mm structure and copy contents from the
* mm structure of the passed in task structure.
*/
struct mm_struct *dup_mm(struct task_struct *tsk)
{
struct mm_struct *mm, *oldmm = current->mm;
int err;
if (!oldmm)
return NULL;
mm = allocate_mm();
if (!mm)
goto fail_nomem;
memcpy(mm, oldmm, sizeof(*mm));
mm_init_cpumask(mm);
/* Initializing for Swap token stuff */
mm->token_priority = 0;
mm->last_interval = 0;
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
mm->pmd_huge_pte = NULL;
#endif
uprobe_reset_state(mm);
if (!mm_init(mm, tsk))
goto fail_nomem;
if (init_new_context(tsk, mm))
goto fail_nocontext;
dup_mm_exe_file(oldmm, mm);
err = dup_mmap(mm, oldmm);
if (err)
goto free_pt;
#ifdef CONFIG_HOMECACHE
{
/* Reset vm_pid on all vmas. In the new mm_struct, we
* want to switch anything that was associated with
* the parent to be associated with the child, and
* clear everything else.
*/
struct vm_area_struct *mpnt;
down_write_nested(&mm->mmap_sem, SINGLE_DEPTH_NESTING);
for (mpnt = mm->mmap; mpnt; mpnt = mpnt->vm_next) {
if (mpnt->vm_pid == current->pid)
mpnt->vm_pid = tsk->pid;
else
mpnt->vm_pid = 0;
}
up_write(&mm->mmap_sem);
}
#endif
mm->hiwater_rss = get_mm_rss(mm);
mm->hiwater_vm = mm->total_vm;
if (mm->binfmt && !try_module_get(mm->binfmt->module))
goto free_pt;
return mm;
free_pt:
/* don't put binfmt in mmput, we haven't got module yet */
mm->binfmt = NULL;
mmput(mm);
fail_nomem:
return NULL;
fail_nocontext:
/*
* If init_new_context() failed, we cannot use mmput() to free the mm
* because it calls destroy_context()
*/
mm_free_pgd(mm);
free_mm(mm);
return NULL;
}
/**
* nilfs_fill_super() - initialize a super block instance
* @sb: super_block
* @data: mount options
* @silent: silent mode flag
*
* This function is called exclusively by nilfs->ns_mount_mutex.
* So, the recovery process is protected from other simultaneous mounts.
*/
static int
nilfs_fill_super(struct super_block *sb, void *data, int silent)
{
struct the_nilfs *nilfs;
struct nilfs_root *fsroot;
__u64 cno;
int err;
nilfs = alloc_nilfs(sb);
if (!nilfs)
return -ENOMEM;
sb->s_fs_info = nilfs;
err = init_nilfs(nilfs, sb, (char *)data);
if (err)
goto failed_nilfs;
sb->s_op = &nilfs_sops;
sb->s_export_op = &nilfs_export_ops;
sb->s_root = NULL;
sb->s_time_gran = 1;
sb->s_max_links = NILFS_LINK_MAX;
sb->s_bdi = &bdev_get_queue(sb->s_bdev)->backing_dev_info;
err = load_nilfs(nilfs, sb);
if (err)
goto failed_nilfs;
cno = nilfs_last_cno(nilfs);
err = nilfs_attach_checkpoint(sb, cno, true, &fsroot);
if (err) {
nilfs_msg(sb, KERN_ERR,
"error %d while loading last checkpoint (checkpoint number=%llu)",
err, (unsigned long long)cno);
goto failed_unload;
}
if (!(sb->s_flags & MS_RDONLY)) {
err = nilfs_attach_log_writer(sb, fsroot);
if (err)
goto failed_checkpoint;
}
err = nilfs_get_root_dentry(sb, fsroot, &sb->s_root);
if (err)
goto failed_segctor;
nilfs_put_root(fsroot);
if (!(sb->s_flags & MS_RDONLY)) {
down_write(&nilfs->ns_sem);
nilfs_setup_super(sb, true);
up_write(&nilfs->ns_sem);
}
return 0;
failed_segctor:
nilfs_detach_log_writer(sb);
failed_checkpoint:
nilfs_put_root(fsroot);
failed_unload:
iput(nilfs->ns_sufile);
iput(nilfs->ns_cpfile);
iput(nilfs->ns_dat);
failed_nilfs:
destroy_nilfs(nilfs);
return err;
}
/*
* Garbage collector for unused keys.
*
* This is done in process context so that we don't have to disable interrupts
* all over the place. key_put() schedules this rather than trying to do the
* cleanup itself, which means key_put() doesn't have to sleep.
*/
static void key_garbage_collector(struct work_struct *work)
{
static LIST_HEAD(graveyard);
static u8 gc_state; /* Internal persistent state */
#define KEY_GC_REAP_AGAIN 0x01 /* - Need another cycle */
#define KEY_GC_REAPING_LINKS 0x02 /* - We need to reap links */
#define KEY_GC_SET_TIMER 0x04 /* - We need to restart the timer */
#define KEY_GC_REAPING_DEAD_1 0x10 /* - We need to mark dead keys */
#define KEY_GC_REAPING_DEAD_2 0x20 /* - We need to reap dead key links */
#define KEY_GC_REAPING_DEAD_3 0x40 /* - We need to reap dead keys */
#define KEY_GC_FOUND_DEAD_KEY 0x80 /* - We found at least one dead key */
struct rb_node *cursor;
struct key *key;
time_t new_timer, limit;
kenter("[%lx,%x]", key_gc_flags, gc_state);
limit = current_kernel_time().tv_sec;
if (limit > key_gc_delay)
limit -= key_gc_delay;
else
limit = key_gc_delay;
/* Work out what we're going to be doing in this pass */
gc_state &= KEY_GC_REAPING_DEAD_1 | KEY_GC_REAPING_DEAD_2;
gc_state <<= 1;
if (test_and_clear_bit(KEY_GC_KEY_EXPIRED, &key_gc_flags))
gc_state |= KEY_GC_REAPING_LINKS | KEY_GC_SET_TIMER;
if (test_and_clear_bit(KEY_GC_REAP_KEYTYPE, &key_gc_flags))
gc_state |= KEY_GC_REAPING_DEAD_1;
kdebug("new pass %x", gc_state);
new_timer = LONG_MAX;
/* As only this function is permitted to remove things from the key
* serial tree, if cursor is non-NULL then it will always point to a
* valid node in the tree - even if lock got dropped.
*/
spin_lock(&key_serial_lock);
cursor = rb_first(&key_serial_tree);
continue_scanning:
while (cursor) {
key = rb_entry(cursor, struct key, serial_node);
cursor = rb_next(cursor);
if (atomic_read(&key->usage) == 0)
goto found_unreferenced_key;
if (unlikely(gc_state & KEY_GC_REAPING_DEAD_1)) {
if (key->type == key_gc_dead_keytype) {
gc_state |= KEY_GC_FOUND_DEAD_KEY;
set_bit(KEY_FLAG_DEAD, &key->flags);
key->perm = 0;
goto skip_dead_key;
}
}
if (gc_state & KEY_GC_SET_TIMER) {
if (key->expiry > limit && key->expiry < new_timer) {
kdebug("will expire %x in %ld",
key_serial(key), key->expiry - limit);
new_timer = key->expiry;
}
}
if (unlikely(gc_state & KEY_GC_REAPING_DEAD_2))
if (key->type == key_gc_dead_keytype)
gc_state |= KEY_GC_FOUND_DEAD_KEY;
if ((gc_state & KEY_GC_REAPING_LINKS) ||
unlikely(gc_state & KEY_GC_REAPING_DEAD_2)) {
if (key->type == &key_type_keyring)
goto found_keyring;
}
if (unlikely(gc_state & KEY_GC_REAPING_DEAD_3))
if (key->type == key_gc_dead_keytype)
goto destroy_dead_key;
skip_dead_key:
if (spin_is_contended(&key_serial_lock) || need_resched())
goto contended;
}
contended:
spin_unlock(&key_serial_lock);
maybe_resched:
if (cursor) {
cond_resched();
spin_lock(&key_serial_lock);
goto continue_scanning;
}
/* We've completed the pass. Set the timer if we need to and queue a
* new cycle if necessary. We keep executing cycles until we find one
* where we didn't reap any keys.
*/
kdebug("pass complete");
if (gc_state & KEY_GC_SET_TIMER && new_timer != (time_t)LONG_MAX) {
new_timer += key_gc_delay;
key_schedule_gc(new_timer);
}
if (unlikely(gc_state & KEY_GC_REAPING_DEAD_2) ||
!list_empty(&graveyard)) {
/* Make sure that all pending keyring payload destructions are
* fulfilled and that people aren't now looking at dead or
* dying keys that they don't have a reference upon or a link
* to.
*/
kdebug("gc sync");
synchronize_rcu();
}
if (!list_empty(&graveyard)) {
kdebug("gc keys");
key_gc_unused_keys(&graveyard);
}
if (unlikely(gc_state & (KEY_GC_REAPING_DEAD_1 |
KEY_GC_REAPING_DEAD_2))) {
if (!(gc_state & KEY_GC_FOUND_DEAD_KEY)) {
/* No remaining dead keys: short circuit the remaining
* keytype reap cycles.
*/
kdebug("dead short");
gc_state &= ~(KEY_GC_REAPING_DEAD_1 | KEY_GC_REAPING_DEAD_2);
gc_state |= KEY_GC_REAPING_DEAD_3;
} else {
gc_state |= KEY_GC_REAP_AGAIN;
}
}
if (unlikely(gc_state & KEY_GC_REAPING_DEAD_3)) {
kdebug("dead wake");
smp_mb();
clear_bit(KEY_GC_REAPING_KEYTYPE, &key_gc_flags);
wake_up_bit(&key_gc_flags, KEY_GC_REAPING_KEYTYPE);
}
if (gc_state & KEY_GC_REAP_AGAIN)
schedule_work(&key_gc_work);
kleave(" [end %x]", gc_state);
return;
/* We found an unreferenced key - once we've removed it from the tree,
* we can safely drop the lock.
*/
found_unreferenced_key:
kdebug("unrefd key %d", key->serial);
rb_erase(&key->serial_node, &key_serial_tree);
spin_unlock(&key_serial_lock);
list_add_tail(&key->graveyard_link, &graveyard);
gc_state |= KEY_GC_REAP_AGAIN;
goto maybe_resched;
/* We found a keyring and we need to check the payload for links to
* dead or expired keys. We don't flag another reap immediately as we
* have to wait for the old payload to be destroyed by RCU before we
* can reap the keys to which it refers.
*/
found_keyring:
spin_unlock(&key_serial_lock);
kdebug("scan keyring %d", key->serial);
key_gc_keyring(key, limit);
goto maybe_resched;
/* We found a dead key that is still referenced. Reset its type and
* destroy its payload with its semaphore held.
*/
destroy_dead_key:
spin_unlock(&key_serial_lock);
kdebug("destroy key %d", key->serial);
down_write(&key->sem);
key->type = &key_type_dead;
if (key_gc_dead_keytype->destroy)
key_gc_dead_keytype->destroy(key);
memset(&key->payload, KEY_DESTROY, sizeof(key->payload));
up_write(&key->sem);
goto maybe_resched;
}
static int nilfs_remount(struct super_block *sb, int *flags, char *data)
{
struct the_nilfs *nilfs = sb->s_fs_info;
unsigned long old_sb_flags;
unsigned long old_mount_opt;
int err;
sync_filesystem(sb);
old_sb_flags = sb->s_flags;
old_mount_opt = nilfs->ns_mount_opt;
if (!parse_options(data, sb, 1)) {
err = -EINVAL;
goto restore_opts;
}
sb->s_flags = (sb->s_flags & ~MS_POSIXACL);
err = -EINVAL;
if (!nilfs_valid_fs(nilfs)) {
nilfs_msg(sb, KERN_WARNING,
"couldn't remount because the filesystem is in an incomplete recovery state");
goto restore_opts;
}
if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY))
goto out;
if (*flags & MS_RDONLY) {
/* Shutting down log writer */
nilfs_detach_log_writer(sb);
sb->s_flags |= MS_RDONLY;
/*
* Remounting a valid RW partition RDONLY, so set
* the RDONLY flag and then mark the partition as valid again.
*/
down_write(&nilfs->ns_sem);
nilfs_cleanup_super(sb);
up_write(&nilfs->ns_sem);
} else {
__u64 features;
struct nilfs_root *root;
/*
* Mounting a RDONLY partition read-write, so reread and
* store the current valid flag. (It may have been changed
* by fsck since we originally mounted the partition.)
*/
down_read(&nilfs->ns_sem);
features = le64_to_cpu(nilfs->ns_sbp[0]->s_feature_compat_ro) &
~NILFS_FEATURE_COMPAT_RO_SUPP;
up_read(&nilfs->ns_sem);
if (features) {
nilfs_msg(sb, KERN_WARNING,
"couldn't remount RDWR because of unsupported optional features (%llx)",
(unsigned long long)features);
err = -EROFS;
goto restore_opts;
}
sb->s_flags &= ~MS_RDONLY;
root = NILFS_I(d_inode(sb->s_root))->i_root;
err = nilfs_attach_log_writer(sb, root);
if (err)
goto restore_opts;
down_write(&nilfs->ns_sem);
nilfs_setup_super(sb, true);
up_write(&nilfs->ns_sem);
}
out:
return 0;
restore_opts:
sb->s_flags = old_sb_flags;
nilfs->ns_mount_opt = old_mount_opt;
return err;
}
int notify_change(struct dentry * dentry, struct iattr * attr)
{
struct inode *inode = dentry->d_inode;
mode_t mode = inode->i_mode;
int error;
struct timespec now;
unsigned int ia_valid = attr->ia_valid;
if (ia_valid & (ATTR_MODE | ATTR_UID | ATTR_GID | ATTR_TIMES_SET)) {
if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
return -EPERM;
}
now = current_fs_time(inode->i_sb);
attr->ia_ctime = now;
if (!(ia_valid & ATTR_ATIME_SET))
attr->ia_atime = now;
if (!(ia_valid & ATTR_MTIME_SET))
attr->ia_mtime = now;
if (ia_valid & ATTR_KILL_PRIV) {
attr->ia_valid &= ~ATTR_KILL_PRIV;
ia_valid &= ~ATTR_KILL_PRIV;
error = security_inode_need_killpriv(dentry);
if (error > 0)
error = security_inode_killpriv(dentry);
if (error)
return error;
}
/*
* We now pass ATTR_KILL_S*ID to the lower level setattr function so
* that the function has the ability to reinterpret a mode change
* that's due to these bits. This adds an implicit restriction that
* no function will ever call notify_change with both ATTR_MODE and
* ATTR_KILL_S*ID set.
*/
if ((ia_valid & (ATTR_KILL_SUID|ATTR_KILL_SGID)) &&
(ia_valid & ATTR_MODE))
BUG();
if (ia_valid & ATTR_KILL_SUID) {
if (mode & S_ISUID) {
ia_valid = attr->ia_valid |= ATTR_MODE;
attr->ia_mode = (inode->i_mode & ~S_ISUID);
}
}
if (ia_valid & ATTR_KILL_SGID) {
if ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP)) {
if (!(ia_valid & ATTR_MODE)) {
ia_valid = attr->ia_valid |= ATTR_MODE;
attr->ia_mode = inode->i_mode;
}
attr->ia_mode &= ~S_ISGID;
}
}
if (!(attr->ia_valid & ~(ATTR_KILL_SUID | ATTR_KILL_SGID)))
return 0;
error = security_inode_setattr(dentry, attr);
if (error)
return error;
if (ia_valid & ATTR_SIZE)
down_write(&dentry->d_inode->i_alloc_sem);
if (inode->i_op->setattr)
error = inode->i_op->setattr(dentry, attr);
else
error = simple_setattr(dentry, attr);
if (ia_valid & ATTR_SIZE)
up_write(&dentry->d_inode->i_alloc_sem);
if (!error)
fsnotify_change(dentry, ia_valid);
return error;
}
/*
* Try to add the new entry to the inline data.
* If succeeds, return 0. If not, extended the inline dir and copied data to
* the new created block.
*/
int ext4_try_add_inline_entry(handle_t *handle, struct dentry *dentry,
struct inode *inode)
{
int ret, inline_size;
void *inline_start;
struct ext4_iloc iloc;
struct inode *dir = dentry->d_parent->d_inode;
ret = ext4_get_inode_loc(dir, &iloc);
if (ret)
return ret;
down_write(&EXT4_I(dir)->xattr_sem);
if (!ext4_has_inline_data(dir))
goto out;
inline_start = (void *)ext4_raw_inode(&iloc)->i_block +
EXT4_INLINE_DOTDOT_SIZE;
inline_size = EXT4_MIN_INLINE_DATA_SIZE - EXT4_INLINE_DOTDOT_SIZE;
ret = ext4_add_dirent_to_inline(handle, dentry, inode, &iloc,
inline_start, inline_size);
if (ret != -ENOSPC)
goto out;
/* check whether it can be inserted to inline xattr space. */
inline_size = EXT4_I(dir)->i_inline_size -
EXT4_MIN_INLINE_DATA_SIZE;
if (!inline_size) {
/* Try to use the xattr space.*/
ret = ext4_update_inline_dir(handle, dir, &iloc);
if (ret && ret != -ENOSPC)
goto out;
inline_size = EXT4_I(dir)->i_inline_size -
EXT4_MIN_INLINE_DATA_SIZE;
}
if (inline_size) {
inline_start = ext4_get_inline_xattr_pos(dir, &iloc);
ret = ext4_add_dirent_to_inline(handle, dentry, inode, &iloc,
inline_start, inline_size);
if (ret != -ENOSPC)
goto out;
}
/*
* The inline space is filled up, so create a new block for it.
* As the extent tree will be created, we have to save the inline
* dir first.
*/
ret = ext4_convert_inline_data_nolock(handle, dir, &iloc);
out:
ext4_mark_inode_dirty(handle, dir);
up_write(&EXT4_I(dir)->xattr_sem);
brelse(iloc.bh);
return ret;
}
/**
* nilfs_fill_super() - initialize a super block instance
* @sb: super_block
* @data: mount options
* @silent: silent mode flag
* @nilfs: the_nilfs struct
*
* This function is called exclusively by bd_mount_mutex.
* So, the recovery process is protected from other simultaneous mounts.
*/
static int
nilfs_fill_super(struct super_block *sb, void *data, int silent,
struct the_nilfs *nilfs)
{
struct nilfs_sb_info *sbi;
struct inode *root;
__u64 cno;
int err;
sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
if (!sbi)
return -ENOMEM;
sb->s_fs_info = sbi;
get_nilfs(nilfs);
sbi->s_nilfs = nilfs;
sbi->s_super = sb;
err = init_nilfs(nilfs, sbi, (char *)data);
if (err)
goto failed_sbi;
spin_lock_init(&sbi->s_inode_lock);
INIT_LIST_HEAD(&sbi->s_dirty_files);
INIT_LIST_HEAD(&sbi->s_list);
/*
* Following initialization is overlapped because
* nilfs_sb_info structure has been cleared at the beginning.
* But we reserve them to keep our interest and make ready
* for the future change.
*/
get_random_bytes(&sbi->s_next_generation,
sizeof(sbi->s_next_generation));
spin_lock_init(&sbi->s_next_gen_lock);
sb->s_op = &nilfs_sops;
sb->s_export_op = &nilfs_export_ops;
sb->s_root = NULL;
sb->s_time_gran = 1;
if (!nilfs_loaded(nilfs)) {
err = load_nilfs(nilfs, sbi);
if (err)
goto failed_sbi;
}
cno = nilfs_last_cno(nilfs);
if (sb->s_flags & MS_RDONLY) {
if (nilfs_test_opt(sbi, SNAPSHOT)) {
err = nilfs_cpfile_is_snapshot(nilfs->ns_cpfile,
sbi->s_snapshot_cno);
if (err < 0)
goto failed_sbi;
if (!err) {
printk(KERN_ERR
"NILFS: The specified checkpoint is "
"not a snapshot "
"(checkpoint number=%llu).\n",
(unsigned long long)sbi->s_snapshot_cno);
err = -EINVAL;
goto failed_sbi;
}
cno = sbi->s_snapshot_cno;
} else
/* Read-only mount */
sbi->s_snapshot_cno = cno;
}
err = nilfs_attach_checkpoint(sbi, cno);
if (err) {
printk(KERN_ERR "NILFS: error loading a checkpoint"
" (checkpoint number=%llu).\n", (unsigned long long)cno);
goto failed_sbi;
}
if (!(sb->s_flags & MS_RDONLY)) {
err = nilfs_attach_segment_constructor(sbi);
if (err)
goto failed_checkpoint;
}
root = nilfs_iget(sb, NILFS_ROOT_INO);
if (IS_ERR(root)) {
printk(KERN_ERR "NILFS: get root inode failed\n");
err = PTR_ERR(root);
goto failed_segctor;
}
if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
iput(root);
printk(KERN_ERR "NILFS: corrupt root inode.\n");
err = -EINVAL;
goto failed_segctor;
}
sb->s_root = d_alloc_root(root);
if (!sb->s_root) {
iput(root);
printk(KERN_ERR "NILFS: get root dentry failed\n");
err = -ENOMEM;
goto failed_segctor;
}
if (!(sb->s_flags & MS_RDONLY)) {
down_write(&nilfs->ns_sem);
nilfs_setup_super(sbi);
up_write(&nilfs->ns_sem);
}
err = nilfs_mark_recovery_complete(sbi);
if (unlikely(err)) {
printk(KERN_ERR "NILFS: recovery failed.\n");
goto failed_root;
}
return 0;
failed_root:
dput(sb->s_root);
sb->s_root = NULL;
failed_segctor:
nilfs_detach_segment_constructor(sbi);
failed_checkpoint:
nilfs_detach_checkpoint(sbi);
failed_sbi:
put_nilfs(nilfs);
sb->s_fs_info = NULL;
kfree(sbi);
return err;
}
static int ext4_convert_inline_data_to_extent(struct address_space *mapping,
struct inode *inode,
unsigned flags)
{
int ret, needed_blocks;
handle_t *handle = NULL;
int retries = 0, sem_held = 0;
struct page *page = NULL;
unsigned from, to;
struct ext4_iloc iloc;
if (!ext4_has_inline_data(inode)) {
/*
* clear the flag so that no new write
* will trap here again.
*/
ext4_clear_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA);
return 0;
}
needed_blocks = ext4_writepage_trans_blocks(inode);
ret = ext4_get_inode_loc(inode, &iloc);
if (ret)
return ret;
retry:
handle = ext4_journal_start(inode, EXT4_HT_WRITE_PAGE, needed_blocks);
if (IS_ERR(handle)) {
ret = PTR_ERR(handle);
handle = NULL;
goto out;
}
/* We cannot recurse into the filesystem as the transaction is already
* started */
flags |= AOP_FLAG_NOFS;
page = grab_cache_page_write_begin(mapping, 0, flags);
if (!page) {
ret = -ENOMEM;
goto out;
}
down_write(&EXT4_I(inode)->xattr_sem);
sem_held = 1;
/* If some one has already done this for us, just exit. */
if (!ext4_has_inline_data(inode)) {
ret = 0;
goto out;
}
from = 0;
to = ext4_get_inline_size(inode);
if (!PageUptodate(page)) {
ret = ext4_read_inline_page(inode, page);
if (ret < 0)
goto out;
}
ret = ext4_destroy_inline_data_nolock(handle, inode);
if (ret)
goto out;
if (ext4_should_dioread_nolock(inode))
ret = __block_write_begin(page, from, to, ext4_get_block_write);
else
ret = __block_write_begin(page, from, to, ext4_get_block);
if (!ret && ext4_should_journal_data(inode)) {
ret = ext4_walk_page_buffers(handle, page_buffers(page),
from, to, NULL,
do_journal_get_write_access);
}
if (ret) {
unlock_page(page);
page_cache_release(page);
page = NULL;
ext4_orphan_add(handle, inode);
up_write(&EXT4_I(inode)->xattr_sem);
sem_held = 0;
ext4_journal_stop(handle);
handle = NULL;
ext4_truncate_failed_write(inode);
/*
* If truncate failed early the inode might
* still be on the orphan list; we need to
* make sure the inode is removed from the
* orphan list in that case.
*/
if (inode->i_nlink)
ext4_orphan_del(NULL, inode);
}
if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
goto retry;
if (page)
block_commit_write(page, from, to);
out:
if (page) {
unlock_page(page);
page_cache_release(page);
}
if (sem_held)
up_write(&EXT4_I(inode)->xattr_sem);
if (handle)
ext4_journal_stop(handle);
brelse(iloc.bh);
return ret;
}
static int nilfs_remount(struct super_block *sb, int *flags, char *data)
{
struct nilfs_sb_info *sbi = NILFS_SB(sb);
struct nilfs_super_block *sbp;
struct the_nilfs *nilfs = sbi->s_nilfs;
unsigned long old_sb_flags;
struct nilfs_mount_options old_opts;
int err;
old_sb_flags = sb->s_flags;
old_opts.mount_opt = sbi->s_mount_opt;
old_opts.snapshot_cno = sbi->s_snapshot_cno;
if (!parse_options(data, sb)) {
err = -EINVAL;
goto restore_opts;
}
sb->s_flags = (sb->s_flags & ~MS_POSIXACL);
if ((*flags & MS_RDONLY) &&
sbi->s_snapshot_cno != old_opts.snapshot_cno) {
printk(KERN_WARNING "NILFS (device %s): couldn't "
"remount to a different snapshot. \n",
sb->s_id);
err = -EINVAL;
goto restore_opts;
}
if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY))
goto out;
if (*flags & MS_RDONLY) {
/* Shutting down the segment constructor */
nilfs_detach_segment_constructor(sbi);
sb->s_flags |= MS_RDONLY;
sbi->s_snapshot_cno = nilfs_last_cno(nilfs);
/* nilfs_set_opt(sbi, SNAPSHOT); */
/*
* Remounting a valid RW partition RDONLY, so set
* the RDONLY flag and then mark the partition as valid again.
*/
down_write(&nilfs->ns_sem);
sbp = nilfs->ns_sbp[0];
if (!(sbp->s_state & le16_to_cpu(NILFS_VALID_FS)) &&
(nilfs->ns_mount_state & NILFS_VALID_FS))
sbp->s_state = cpu_to_le16(nilfs->ns_mount_state);
sbp->s_mtime = cpu_to_le64(get_seconds());
nilfs_commit_super(sbi, 1);
up_write(&nilfs->ns_sem);
} else {
/*
* Mounting a RDONLY partition read-write, so reread and
* store the current valid flag. (It may have been changed
* by fsck since we originally mounted the partition.)
*/
down(&sb->s_bdev->bd_mount_sem);
/* Check existing RW-mount */
if (test_exclusive_mount(sb->s_type, sb->s_bdev, 0)) {
printk(KERN_WARNING "NILFS (device %s): couldn't "
"remount because a RW-mount exists.\n",
sb->s_id);
err = -EBUSY;
goto rw_remount_failed;
}
if (sbi->s_snapshot_cno != nilfs_last_cno(nilfs)) {
printk(KERN_WARNING "NILFS (device %s): couldn't "
"remount because the current RO-mount is not "
"the latest one.\n",
sb->s_id);
err = -EINVAL;
goto rw_remount_failed;
}
sb->s_flags &= ~MS_RDONLY;
nilfs_clear_opt(sbi, SNAPSHOT);
sbi->s_snapshot_cno = 0;
err = nilfs_attach_segment_constructor(sbi);
if (err)
goto rw_remount_failed;
down_write(&nilfs->ns_sem);
nilfs_setup_super(sbi);
up_write(&nilfs->ns_sem);
up(&sb->s_bdev->bd_mount_sem);
}
out:
return 0;
rw_remount_failed:
up(&sb->s_bdev->bd_mount_sem);
restore_opts:
sb->s_flags = old_sb_flags;
sbi->s_mount_opt = old_opts.mount_opt;
sbi->s_snapshot_cno = old_opts.snapshot_cno;
return err;
}
/*
* We need to be able to change a mapping table under a mounted
* filesystem. For example we might want to move some data in
* the background. Before the table can be swapped with
* dm_bind_table, dm_suspend must be called to flush any in
* flight bios and ensure that any further io gets deferred.
*/
int dm_suspend(struct mapped_device *md)
{
struct dm_table *map;
DECLARE_WAITQUEUE(wait, current);
/* Flush I/O to the device. */
down_read(&md->lock);
if (test_bit(DMF_BLOCK_IO, &md->flags)) {
up_read(&md->lock);
return -EINVAL;
}
map = dm_get_table(md);
if (map)
dm_table_presuspend_targets(map);
__lock_fs(md);
up_read(&md->lock);
/*
* First we set the BLOCK_IO flag so no more ios will be
* mapped.
*/
down_write(&md->lock);
if (test_bit(DMF_BLOCK_IO, &md->flags)) {
/*
* If we get here we know another thread is
* trying to suspend as well, so we leave the fs
* locked for this thread.
*/
up_write(&md->lock);
return -EINVAL;
}
set_bit(DMF_BLOCK_IO, &md->flags);
add_wait_queue(&md->wait, &wait);
up_write(&md->lock);
/* unplug */
if (map) {
dm_table_unplug_all(map);
dm_table_put(map);
}
/*
* Then we wait for the already mapped ios to
* complete.
*/
while (1) {
set_current_state(TASK_INTERRUPTIBLE);
if (!atomic_read(&md->pending) || signal_pending(current))
break;
io_schedule();
}
set_current_state(TASK_RUNNING);
down_write(&md->lock);
remove_wait_queue(&md->wait, &wait);
/* were we interrupted ? */
if (atomic_read(&md->pending)) {
__unlock_fs(md);
clear_bit(DMF_BLOCK_IO, &md->flags);
up_write(&md->lock);
return -EINTR;
}
set_bit(DMF_SUSPENDED, &md->flags);
map = dm_get_table(md);
if (map)
dm_table_postsuspend_targets(map);
dm_table_put(map);
up_write(&md->lock);
return 0;
}
int
osi_UFSTruncate(struct osi_file *afile, afs_int32 asize)
{
afs_int32 code;
struct osi_stat tstat;
struct iattr newattrs;
struct inode *inode = OSIFILE_INODE(afile);
AFS_STATCNT(osi_Truncate);
/* This routine only shrinks files, and most systems
* have very slow truncates, even when the file is already
* small enough. Check now and save some time.
*/
code = afs_osi_Stat(afile, &tstat);
if (code || tstat.size <= asize)
return code;
ObtainWriteLock(&afs_xosi, 321);
AFS_GUNLOCK();
#ifdef STRUCT_INODE_HAS_I_ALLOC_SEM
down_write(&inode->i_alloc_sem);
#endif
#ifdef STRUCT_INODE_HAS_I_MUTEX
mutex_lock(&inode->i_mutex);
#else
down(&inode->i_sem);
#endif
newattrs.ia_size = asize;
newattrs.ia_valid = ATTR_SIZE | ATTR_CTIME;
#if defined(AFS_LINUX24_ENV)
newattrs.ia_ctime = CURRENT_TIME;
/* avoid notify_change() since it wants to update dentry->d_parent */
lock_kernel();
code = inode_change_ok(inode, &newattrs);
if (!code) {
#ifdef INODE_SETATTR_NOT_VOID
code = inode_setattr(inode, &newattrs);
#else
inode_setattr(inode, &newattrs);
#endif
}
unlock_kernel();
if (!code)
truncate_inode_pages(&inode->i_data, asize);
#else
i_size_write(inode, asize);
if (inode->i_sb->s_op && inode->i_sb->s_op->notify_change) {
code = inode->i_sb->s_op->notify_change(&afile->dentry, &newattrs);
}
if (!code) {
truncate_inode_pages(inode, asize);
if (inode->i_op && inode->i_op->truncate)
inode->i_op->truncate(inode);
}
#endif
code = -code;
#ifdef STRUCT_INODE_HAS_I_MUTEX
mutex_unlock(&inode->i_mutex);
#else
up(&inode->i_sem);
#endif
#ifdef STRUCT_INODE_HAS_I_ALLOC_SEM
up_write(&inode->i_alloc_sem);
#endif
AFS_GLOCK();
ReleaseWriteLock(&afs_xosi);
return code;
}
struct dentry *f2fs_get_parent(struct dentry *child)
{
struct qstr dotdot = {.len = 2, .name = ".."};
unsigned long ino = f2fs_inode_by_name(child->d_inode, &dotdot);
if (!ino)
return ERR_PTR(-ENOENT);
return d_obtain_alias(f2fs_iget(child->d_inode->i_sb, ino));
}
static struct dentry *f2fs_lookup(struct inode *dir, struct dentry *dentry,
struct nameidata *nd)
{
struct inode *inode = NULL;
struct f2fs_dir_entry *de;
struct page *page;
if (dentry->d_name.len > F2FS_NAME_LEN)
return ERR_PTR(-ENAMETOOLONG);
de = f2fs_find_entry(dir, &dentry->d_name, &page, nd ? nd->flags : 0);
if (de) {
nid_t ino = le32_to_cpu(de->ino);
if (!f2fs_has_inline_dentry(dir))
kunmap(page);
f2fs_put_page(page, 0);
inode = f2fs_iget(dir->i_sb, ino);
if (IS_ERR(inode))
return ERR_CAST(inode);
}
return d_splice_alias(inode, dentry);
}
static int f2fs_unlink(struct inode *dir, struct dentry *dentry)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
struct inode *inode = dentry->d_inode;
struct f2fs_dir_entry *de;
struct page *page;
int err = -ENOENT;
trace_f2fs_unlink_enter(dir, dentry);
f2fs_balance_fs(sbi);
de = f2fs_find_entry(dir, &dentry->d_name, &page, 0);
if (!de)
goto fail;
f2fs_lock_op(sbi);
err = acquire_orphan_inode(sbi);
if (err) {
f2fs_unlock_op(sbi);
if (!f2fs_has_inline_dentry(dir))
kunmap(page);
f2fs_put_page(page, 0);
goto fail;
}
f2fs_delete_entry(de, page, dir, inode);
f2fs_unlock_op(sbi);
/* In order to evict this inode, we set it dirty */
mark_inode_dirty(inode);
fail:
trace_f2fs_unlink_exit(inode, err);
return err;
}
static int f2fs_symlink(struct inode *dir, struct dentry *dentry,
const char *symname)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
struct inode *inode;
size_t symlen = strlen(symname) + 1;
int err;
f2fs_balance_fs(sbi);
inode = f2fs_new_inode(dir, S_IFLNK | S_IRWXUGO);
if (IS_ERR(inode))
return PTR_ERR(inode);
inode->i_op = &f2fs_symlink_inode_operations;
inode->i_mapping->a_ops = &f2fs_dblock_aops;
f2fs_lock_op(sbi);
err = f2fs_add_link(dentry, inode);
if (err)
goto out;
f2fs_unlock_op(sbi);
err = page_symlink(inode, symname, symlen);
alloc_nid_done(sbi, inode->i_ino);
d_instantiate(dentry, inode);
unlock_new_inode(inode);
return err;
out:
handle_failed_inode(inode);
return err;
}
static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
struct inode *inode;
int err;
f2fs_balance_fs(sbi);
inode = f2fs_new_inode(dir, S_IFDIR | mode);
if (IS_ERR(inode))
return PTR_ERR(inode);
inode->i_op = &f2fs_dir_inode_operations;
inode->i_fop = &f2fs_dir_operations;
inode->i_mapping->a_ops = &f2fs_dblock_aops;
mapping_set_gfp_mask(inode->i_mapping, GFP_F2FS_ZERO);
set_inode_flag(F2FS_I(inode), FI_INC_LINK);
f2fs_lock_op(sbi);
err = f2fs_add_link(dentry, inode);
if (err)
goto out_fail;
f2fs_unlock_op(sbi);
stat_inc_inline_dir(inode);
alloc_nid_done(sbi, inode->i_ino);
d_instantiate(dentry, inode);
unlock_new_inode(inode);
return 0;
out_fail:
clear_inode_flag(F2FS_I(inode), FI_INC_LINK);
handle_failed_inode(inode);
return err;
}
static int f2fs_rmdir(struct inode *dir, struct dentry *dentry)
{
struct inode *inode = dentry->d_inode;
if (f2fs_empty_dir(inode))
return f2fs_unlink(dir, dentry);
return -ENOTEMPTY;
}
static int f2fs_mknod(struct inode *dir, struct dentry *dentry,
umode_t mode, dev_t rdev)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
struct inode *inode;
int err = 0;
if (!new_valid_dev(rdev))
return -EINVAL;
f2fs_balance_fs(sbi);
inode = f2fs_new_inode(dir, mode);
if (IS_ERR(inode))
return PTR_ERR(inode);
init_special_inode(inode, inode->i_mode, rdev);
inode->i_op = &f2fs_special_inode_operations;
f2fs_lock_op(sbi);
err = f2fs_add_link(dentry, inode);
if (err)
goto out;
f2fs_unlock_op(sbi);
alloc_nid_done(sbi, inode->i_ino);
d_instantiate(dentry, inode);
unlock_new_inode(inode);
return 0;
out:
handle_failed_inode(inode);
return err;
}
static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
struct inode *new_dir, struct dentry *new_dentry)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(old_dir);
struct inode *old_inode = old_dentry->d_inode;
struct inode *new_inode = new_dentry->d_inode;
struct page *old_dir_page;
struct page *old_page, *new_page;
struct f2fs_dir_entry *old_dir_entry = NULL;
struct f2fs_dir_entry *old_entry;
struct f2fs_dir_entry *new_entry;
int err = -ENOENT;
f2fs_balance_fs(sbi);
old_entry = f2fs_find_entry(old_dir, &old_dentry->d_name, &old_page, 0);
if (!old_entry)
goto out;
if (S_ISDIR(old_inode->i_mode)) {
err = -EIO;
old_dir_entry = f2fs_parent_dir(old_inode, &old_dir_page);
if (!old_dir_entry)
goto out_old;
}
if (new_inode) {
err = -ENOTEMPTY;
if (old_dir_entry && !f2fs_empty_dir(new_inode))
goto out_dir;
err = -ENOENT;
new_entry = f2fs_find_entry(new_dir, &new_dentry->d_name,
&new_page, 0);
if (!new_entry)
goto out_dir;
f2fs_lock_op(sbi);
err = acquire_orphan_inode(sbi);
if (err)
goto put_out_dir;
if (update_dent_inode(old_inode, &new_dentry->d_name)) {
release_orphan_inode(sbi);
goto put_out_dir;
}
f2fs_set_link(new_dir, new_entry, new_page, old_inode);
new_inode->i_ctime = CURRENT_TIME;
down_write(&F2FS_I(new_inode)->i_sem);
if (old_dir_entry)
drop_nlink(new_inode);
drop_nlink(new_inode);
up_write(&F2FS_I(new_inode)->i_sem);
mark_inode_dirty(new_inode);
if (!new_inode->i_nlink)
add_orphan_inode(sbi, new_inode->i_ino);
else
release_orphan_inode(sbi);
update_inode_page(old_inode);
update_inode_page(new_inode);
} else {
f2fs_lock_op(sbi);
err = f2fs_add_link(new_dentry, old_inode);
if (err) {
f2fs_unlock_op(sbi);
goto out_dir;
}
if (old_dir_entry) {
inc_nlink(new_dir);
update_inode_page(new_dir);
}
}
down_write(&F2FS_I(old_inode)->i_sem);
file_lost_pino(old_inode);
up_write(&F2FS_I(old_inode)->i_sem);
old_inode->i_ctime = CURRENT_TIME;
mark_inode_dirty(old_inode);
f2fs_delete_entry(old_entry, old_page, old_dir, NULL);
if (old_dir_entry) {
if (old_dir != new_dir) {
f2fs_set_link(old_inode, old_dir_entry,
old_dir_page, new_dir);
update_inode_page(old_inode);
} else {
if (!f2fs_has_inline_dentry(old_inode))
kunmap(old_dir_page);
f2fs_put_page(old_dir_page, 0);
}
drop_nlink(old_dir);
mark_inode_dirty(old_dir);
update_inode_page(old_dir);
}
f2fs_unlock_op(sbi);
return 0;
put_out_dir:
f2fs_unlock_op(sbi);
if (!f2fs_has_inline_dentry(new_dir))
kunmap(new_page);
f2fs_put_page(new_page, 0);
out_dir:
if (old_dir_entry) {
if (!f2fs_has_inline_dentry(old_inode))
kunmap(old_dir_page);
f2fs_put_page(old_dir_page, 0);
}
out_old:
if (!f2fs_has_inline_dentry(old_dir))
kunmap(old_page);
f2fs_put_page(old_page, 0);
out:
return err;
}
const struct inode_operations f2fs_dir_inode_operations = {
.create = f2fs_create,
.lookup = f2fs_lookup,
.link = f2fs_link,
.unlink = f2fs_unlink,
.symlink = f2fs_symlink,
.mkdir = f2fs_mkdir,
.rmdir = f2fs_rmdir,
.mknod = f2fs_mknod,
.rename = f2fs_rename,
.getattr = f2fs_getattr,
.setattr = f2fs_setattr,
.get_acl = f2fs_get_acl,
#ifdef CONFIG_F2FS_FS_XATTR
.setxattr = generic_setxattr,
.getxattr = generic_getxattr,
.listxattr = f2fs_listxattr,
.removexattr = generic_removexattr,
#endif
};
const struct inode_operations f2fs_symlink_inode_operations = {
.readlink = generic_readlink,
.follow_link = page_follow_link_light,
.put_link = page_put_link,
.getattr = f2fs_getattr,
.setattr = f2fs_setattr,
#ifdef CONFIG_F2FS_FS_XATTR
.setxattr = generic_setxattr,
.getxattr = generic_getxattr,
.listxattr = f2fs_listxattr,
.removexattr = generic_removexattr,
#endif
};
const struct inode_operations f2fs_special_inode_operations = {
.getattr = f2fs_getattr,
.setattr = f2fs_setattr,
.get_acl = f2fs_get_acl,
#ifdef CONFIG_F2FS_FS_XATTR
.setxattr = generic_setxattr,
.getxattr = generic_getxattr,
.listxattr = f2fs_listxattr,
.removexattr = generic_removexattr,
#endif
};
static int
islpci_reset_if(islpci_private *priv)
{
long remaining;
int result = -ETIME;
int count;
DEFINE_WAIT(wait);
prepare_to_wait(&priv->reset_done, &wait, TASK_UNINTERRUPTIBLE);
/* now the last step is to reset the interface */
isl38xx_interface_reset(priv->device_base, priv->device_host_address);
islpci_set_state(priv, PRV_STATE_PREINIT);
for(count = 0; count < 2 && result; count++) {
/* The software reset acknowledge needs about 220 msec here.
* Be conservative and wait for up to one second. */
set_current_state(TASK_UNINTERRUPTIBLE);
remaining = schedule_timeout(HZ);
if(remaining > 0) {
result = 0;
break;
}
/* If we're here it's because our IRQ hasn't yet gone through.
* Retry a bit more...
*/
printk(KERN_ERR "%s: no 'reset complete' IRQ seen - retrying\n",
priv->ndev->name);
}
finish_wait(&priv->reset_done, &wait);
if (result) {
printk(KERN_ERR "%s: interface reset failure\n", priv->ndev->name);
return result;
}
islpci_set_state(priv, PRV_STATE_INIT);
/* Now that the device is 100% up, let's allow
* for the other interrupts --
* NOTE: this is not *yet* true since we've only allowed the
* INIT interrupt on the IRQ line. We can perhaps poll
* the IRQ line until we know for sure the reset went through */
isl38xx_enable_common_interrupts(priv->device_base);
down_write(&priv->mib_sem);
result = mgt_commit(priv);
if (result) {
printk(KERN_ERR "%s: interface reset failure\n", priv->ndev->name);
up_write(&priv->mib_sem);
return result;
}
up_write(&priv->mib_sem);
islpci_set_state(priv, PRV_STATE_READY);
printk(KERN_DEBUG "%s: interface reset complete\n", priv->ndev->name);
return 0;
}
void ocfs2_clear_inode(struct inode *inode)
{
int status;
struct ocfs2_inode_info *oi = OCFS2_I(inode);
mlog_entry_void();
if (!inode)
goto bail;
mlog(0, "Clearing inode: %llu, nlink = %u\n",
(unsigned long long)OCFS2_I(inode)->ip_blkno, inode->i_nlink);
mlog_bug_on_msg(OCFS2_SB(inode->i_sb) == NULL,
"Inode=%lu\n", inode->i_ino);
/* To preven remote deletes we hold open lock before, now it
* is time to unlock PR and EX open locks. */
ocfs2_open_unlock(inode);
/* Do these before all the other work so that we don't bounce
* the downconvert thread while waiting to destroy the locks. */
ocfs2_mark_lockres_freeing(&oi->ip_rw_lockres);
ocfs2_mark_lockres_freeing(&oi->ip_inode_lockres);
ocfs2_mark_lockres_freeing(&oi->ip_open_lockres);
/* We very well may get a clear_inode before all an inodes
* metadata has hit disk. Of course, we can't drop any cluster
* locks until the journal has finished with it. The only
* exception here are successfully wiped inodes - their
* metadata can now be considered to be part of the system
* inodes from which it came. */
if (!(OCFS2_I(inode)->ip_flags & OCFS2_INODE_DELETED))
ocfs2_checkpoint_inode(inode);
mlog_bug_on_msg(!list_empty(&oi->ip_io_markers),
"Clear inode of %llu, inode has io markers\n",
(unsigned long long)oi->ip_blkno);
ocfs2_extent_map_trunc(inode, 0);
status = ocfs2_drop_inode_locks(inode);
if (status < 0)
mlog_errno(status);
ocfs2_lock_res_free(&oi->ip_rw_lockres);
ocfs2_lock_res_free(&oi->ip_inode_lockres);
ocfs2_lock_res_free(&oi->ip_open_lockres);
ocfs2_metadata_cache_purge(inode);
mlog_bug_on_msg(oi->ip_metadata_cache.ci_num_cached,
"Clear inode of %llu, inode has %u cache items\n",
(unsigned long long)oi->ip_blkno, oi->ip_metadata_cache.ci_num_cached);
mlog_bug_on_msg(!(oi->ip_flags & OCFS2_INODE_CACHE_INLINE),
"Clear inode of %llu, inode has a bad flag\n",
(unsigned long long)oi->ip_blkno);
mlog_bug_on_msg(spin_is_locked(&oi->ip_lock),
"Clear inode of %llu, inode is locked\n",
(unsigned long long)oi->ip_blkno);
mlog_bug_on_msg(!mutex_trylock(&oi->ip_io_mutex),
"Clear inode of %llu, io_mutex is locked\n",
(unsigned long long)oi->ip_blkno);
mutex_unlock(&oi->ip_io_mutex);
/*
* down_trylock() returns 0, down_write_trylock() returns 1
* kernel 1, world 0
*/
mlog_bug_on_msg(!down_write_trylock(&oi->ip_alloc_sem),
"Clear inode of %llu, alloc_sem is locked\n",
(unsigned long long)oi->ip_blkno);
up_write(&oi->ip_alloc_sem);
mlog_bug_on_msg(oi->ip_open_count,
"Clear inode of %llu has open count %d\n",
(unsigned long long)oi->ip_blkno, oi->ip_open_count);
/* Clear all other flags. */
oi->ip_flags = OCFS2_INODE_CACHE_INLINE;
oi->ip_created_trans = 0;
oi->ip_last_trans = 0;
oi->ip_dir_start_lookup = 0;
oi->ip_blkno = 0ULL;
/*
* ip_jinode is used to track txns against this inode. We ensure that
* the journal is flushed before journal shutdown. Thus it is safe to
* have inodes get cleaned up after journal shutdown.
*/
jbd2_journal_release_jbd_inode(OCFS2_SB(inode->i_sb)->journal->j_journal,
&oi->ip_jinode);
bail:
mlog_exit_void();
}
static void serio_release_driver(struct serio *serio)
{
down_write(&serio_bus.subsys.rwsem);
device_release_driver(&serio->dev);
up_write(&serio_bus.subsys.rwsem);
}
/* SunOS is completely broken... it returns 0 on success, otherwise
* ENOMEM. For sys_sbrk() it wants the old brk value as a return
* on success and ENOMEM as before on failure.
*/
asmlinkage int sunos_brk(unsigned long brk)
{
int freepages, retval = -ENOMEM;
unsigned long rlim;
unsigned long newbrk, oldbrk;
down_write(¤t->mm->mmap_sem);
if (ARCH_SUN4C_SUN4) {
if (brk >= 0x20000000 && brk < 0xe0000000) {
goto out;
}
}
if (brk < current->mm->end_code)
goto out;
newbrk = PAGE_ALIGN(brk);
oldbrk = PAGE_ALIGN(current->mm->brk);
retval = 0;
if (oldbrk == newbrk) {
current->mm->brk = brk;
goto out;
}
/*
* Always allow shrinking brk
*/
if (brk <= current->mm->brk) {
current->mm->brk = brk;
do_munmap(current->mm, newbrk, oldbrk-newbrk);
goto out;
}
/*
* Check against rlimit and stack..
*/
retval = -ENOMEM;
rlim = current->rlim[RLIMIT_DATA].rlim_cur;
if (rlim >= RLIM_INFINITY)
rlim = ~0;
if (brk - current->mm->end_code > rlim)
goto out;
/*
* Check against existing mmap mappings.
*/
if (find_vma_intersection(current->mm, oldbrk, newbrk+PAGE_SIZE))
goto out;
/*
* stupid algorithm to decide if we have enough memory: while
* simple, it hopefully works in most obvious cases.. Easy to
* fool it, but this should catch most mistakes.
*/
freepages = get_page_cache_size();
freepages >>= 1;
freepages += nr_free_pages();
freepages += nr_swap_pages;
freepages -= num_physpages >> 4;
freepages -= (newbrk-oldbrk) >> PAGE_SHIFT;
if (freepages < 0)
goto out;
/*
* Ok, we have probably got enough memory - let it rip.
*/
current->mm->brk = brk;
do_brk(oldbrk, newbrk-oldbrk);
retval = 0;
out:
up_write(¤t->mm->mmap_sem);
return retval;
}
/* This is really VOP_SETATTR() in sheep's clothing */
int
vnode_iop_notify_change(
DENT_T *dent_p,
struct iattr * iattr_p
)
{
VNODE_T *vp;
VATTR_T *vap;
VNODE_T *cvp;
int err = 0;
DENT_T *rdent;
CALL_DATA_T cd;
#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,36)
mdki_boolean_t tooksem = FALSE;
#endif
if (iattr_p->ia_valid & ATTR_SIZE) {
ASSERT_I_SEM_MINE(dent_p->d_inode);
}
if (MDKI_INOISMVFS(dent_p->d_inode)) {
vap = VATTR_ALLOC();
if (vap != NULL) {
vnode_iop_iattr2vattr(iattr_p, vap);
/* reject attempts to use setattr to change object type */
vap->va_mask &= ~AT_TYPE;
mdki_linux_init_call_data(&cd);
vp = ITOV(dent_p->d_inode);
err = VOP_SETATTR(vp, vap, 0, &cd);
err = mdki_errno_unix_to_linux(err);
/* Any underlying cleartxt got its inode truncated via changeattr
* if there's a need to change its size.
*/
if (!err)
mdki_linux_vattr_pullup(vp, vap, vap->va_mask);
VATTR_FREE(vap);
mdki_linux_destroy_call_data(&cd);
} else {
err = -ENOMEM;
}
} else {
rdent = REALDENTRY_LOCKED(dent_p, &cvp);
VNODE_DGET(rdent);
if (rdent && rdent->d_inode) {
#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,36)
err = inode_setattr(dent_p->d_inode, iattr_p);
if (err == 0) {
if (iattr_p->ia_valid & ATTR_SIZE) {
LOCK_INODE(rdent->d_inode);
#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,13)
#if !defined RHEL_UPDATE || RHEL_UPDATE < 5
down_write(&rdent->d_inode->i_alloc_sem);
#endif
#endif
/*
* be paranoid and record the 'taken'ness in case
* the called function squashes ia_valid (as is
* done in nfs_setattr).
*/
tooksem = TRUE;
}
err = MDKI_NOTIFY_CHANGE(rdent, CVN_TO_VFSMNT(cvp), iattr_p);
if (tooksem) {
#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,13)
#if !defined(RHEL_UPDATE) || RHEL_UPDATE < 5
up_write(&rdent->d_inode->i_alloc_sem);
#endif
#endif
UNLOCK_INODE(rdent->d_inode);
}
}
#else /* LINUX_VERSION_CODE < KERNEL_VERSION(2,6,36) */
err = simple_setattr(dent_p, iattr_p);
if (err == 0)
err = MDKI_NOTIFY_CHANGE(rdent, CVN_TO_VFSMNT(cvp), iattr_p);
#endif /* else LINUX_VERSION_CODE < KERNEL_VERSION(2,6,36) */
} else {
/* It looks as though someone removed the realdentry on us.
* I am not sure why this should happen.
*/
err = -ENOENT;
}
if (rdent) {
VNODE_DPUT(rdent);
REALDENTRY_UNLOCK(dent_p, cvp);
}
}
return err;
}