linux-stable/fs/romfs/super.c
Linus Torvalds 511fb5bafe v6.6-vfs.super
-----BEGIN PGP SIGNATURE-----
 
 iHUEABYKAB0WIQRAhzRXHqcMeLMyaSiRxhvAZXjcogUCZOXpbgAKCRCRxhvAZXjc
 oi8PAQCtXelGZHmTcmevsO8p4Qz7hFpkonZ/TnxKf+RdnlNgPgD+NWi+LoRBpaAj
 xk4z8SqJaTTP4WXrG5JZ6o7EQkUL8gE=
 =2e9I
 -----END PGP SIGNATURE-----

Merge tag 'v6.6-vfs.super' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs

Pull superblock updates from Christian Brauner:
 "This contains the super rework that was ready for this cycle. The
  first part changes the order of how we open block devices and allocate
  superblocks, contains various cleanups, simplifications, and a new
  mechanism to wait on superblock state changes.

  This unblocks work to ultimately limit the number of writers to a
  block device. Jan has already scheduled follow-up work that will be
  ready for v6.7 and allows us to restrict the number of writers to a
  given block device. That series builds on this work right here.

  The second part contains filesystem freezing updates.

  Overview:

  The generic superblock changes are rougly organized as follows
  (ignoring additional minor cleanups):

   (1) Removal of the bd_super member from struct block_device.

       This was a very odd back pointer to struct super_block with
       unclear rules. For all relevant places we have other means to get
       the same information so just get rid of this.

   (2) Simplify rules for superblock cleanup.

       Roughly, everything that is allocated during fs_context
       initialization and that's stored in fs_context->s_fs_info needs
       to be cleaned up by the fs_context->free() implementation before
       the superblock allocation function has been called successfully.

       After sget_fc() returned fs_context->s_fs_info has been
       transferred to sb->s_fs_info at which point sb->kill_sb() if
       fully responsible for cleanup. Adhering to these rules means that
       cleanup of sb->s_fs_info in fill_super() is to be avoided as it's
       brittle and inconsistent.

       Cleanup shouldn't be duplicated between sb->put_super() as
       sb->put_super() is only called if sb->s_root has been set aka
       when the filesystem has been successfully born (SB_BORN). That
       complexity should be avoided.

       This also means that block devices are to be closed in
       sb->kill_sb() instead of sb->put_super(). More details in the
       lower section.

   (3) Make it possible to lookup or create a superblock before opening
       block devices

       There's a subtle dependency on (2) as some filesystems did rely
       on fill_super() to be called in order to correctly clean up
       sb->s_fs_info. All these filesystems have been fixed.

   (4) Switch most filesystem to follow the same logic as the generic
       mount code now does as outlined in (3).

   (5) Use the superblock as the holder of the block device. We can now
       easily go back from block device to owning superblock.

   (6) Export and extend the generic fs_holder_ops and use them as
       holder ops everywhere and remove the filesystem specific holder
       ops.

   (7) Call from the block layer up into the filesystem layer when the
       block device is removed, allowing to shut down the filesystem
       without risk of deadlocks.

   (8) Get rid of get_super().

       We can now easily go back from the block device to owning
       superblock and can call up from the block layer into the
       filesystem layer when the device is removed. So no need to wade
       through all registered superblock to find the owning superblock
       anymore"

Link: https://lore.kernel.org/lkml/20230824-prall-intakt-95dbffdee4a0@brauner/

* tag 'v6.6-vfs.super' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (47 commits)
  super: use higher-level helper for {freeze,thaw}
  super: wait until we passed kill super
  super: wait for nascent superblocks
  super: make locking naming consistent
  super: use locking helpers
  fs: simplify invalidate_inodes
  fs: remove get_super
  block: call into the file system for ioctl BLKFLSBUF
  block: call into the file system for bdev_mark_dead
  block: consolidate __invalidate_device and fsync_bdev
  block: drop the "busy inodes on changed media" log message
  dasd: also call __invalidate_device when setting the device offline
  amiflop: don't call fsync_bdev in FDFMTBEG
  floppy: call disk_force_media_change when changing the format
  block: simplify the disk_force_media_change interface
  nbd: call blk_mark_disk_dead in nbd_clear_sock_ioctl
  xfs use fs_holder_ops for the log and RT devices
  xfs: drop s_umount over opening the log and RT devices
  ext4: use fs_holder_ops for the log device
  ext4: drop s_umount over opening the log device
  ...
2023-08-28 11:04:18 -07:00

670 lines
15 KiB
C

/* Block- or MTD-based romfs
*
* Copyright © 2007 Red Hat, Inc. All Rights Reserved.
* Written by David Howells (dhowells@redhat.com)
*
* Derived from: ROMFS file system, Linux implementation
*
* Copyright © 1997-1999 Janos Farkas <chexum@shadow.banki.hu>
*
* Using parts of the minix filesystem
* Copyright © 1991, 1992 Linus Torvalds
*
* and parts of the affs filesystem additionally
* Copyright © 1993 Ray Burr
* Copyright © 1996 Hans-Joachim Widmaier
*
* Changes
* Changed for 2.1.19 modules
* Jan 1997 Initial release
* Jun 1997 2.1.43+ changes
* Proper page locking in read_folio
* Changed to work with 2.1.45+ fs
* Jul 1997 Fixed follow_link
* 2.1.47
* lookup shouldn't return -ENOENT
* from Horst von Brand:
* fail on wrong checksum
* double unlock_super was possible
* correct namelen for statfs
* spotted by Bill Hawes:
* readlink shouldn't iput()
* Jun 1998 2.1.106 from Avery Pennarun: glibc scandir()
* exposed a problem in readdir
* 2.1.107 code-freeze spellchecker run
* Aug 1998 2.1.118+ VFS changes
* Sep 1998 2.1.122 another VFS change (follow_link)
* Apr 1999 2.2.7 no more EBADF checking in
* lookup/readdir, use ERR_PTR
* Jun 1999 2.3.6 d_alloc_root use changed
* 2.3.9 clean up usage of ENOENT/negative
* dentries in lookup
* clean up page flags setting
* (error, uptodate, locking) in
* in read_folio
* use init_special_inode for
* fifos/sockets (and streamline) in
* read_inode, fix _ops table order
* Aug 1999 2.3.16 __initfunc() => __init change
* Oct 1999 2.3.24 page->owner hack obsoleted
* Nov 1999 2.3.27 2.3.25+ page->offset => index change
*
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public Licence
* as published by the Free Software Foundation; either version
* 2 of the Licence, or (at your option) any later version.
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/module.h>
#include <linux/string.h>
#include <linux/fs.h>
#include <linux/time.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/blkdev.h>
#include <linux/fs_context.h>
#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/statfs.h>
#include <linux/mtd/super.h>
#include <linux/ctype.h>
#include <linux/highmem.h>
#include <linux/pagemap.h>
#include <linux/uaccess.h>
#include <linux/major.h>
#include "internal.h"
static struct kmem_cache *romfs_inode_cachep;
static const umode_t romfs_modemap[8] = {
0, /* hard link */
S_IFDIR | 0644, /* directory */
S_IFREG | 0644, /* regular file */
S_IFLNK | 0777, /* symlink */
S_IFBLK | 0600, /* blockdev */
S_IFCHR | 0600, /* chardev */
S_IFSOCK | 0644, /* socket */
S_IFIFO | 0644 /* FIFO */
};
static const unsigned char romfs_dtype_table[] = {
DT_UNKNOWN, DT_DIR, DT_REG, DT_LNK, DT_BLK, DT_CHR, DT_SOCK, DT_FIFO
};
static struct inode *romfs_iget(struct super_block *sb, unsigned long pos);
/*
* read a page worth of data from the image
*/
static int romfs_read_folio(struct file *file, struct folio *folio)
{
struct page *page = &folio->page;
struct inode *inode = page->mapping->host;
loff_t offset, size;
unsigned long fillsize, pos;
void *buf;
int ret;
buf = kmap(page);
if (!buf)
return -ENOMEM;
/* 32 bit warning -- but not for us :) */
offset = page_offset(page);
size = i_size_read(inode);
fillsize = 0;
ret = 0;
if (offset < size) {
size -= offset;
fillsize = size > PAGE_SIZE ? PAGE_SIZE : size;
pos = ROMFS_I(inode)->i_dataoffset + offset;
ret = romfs_dev_read(inode->i_sb, pos, buf, fillsize);
if (ret < 0) {
SetPageError(page);
fillsize = 0;
ret = -EIO;
}
}
if (fillsize < PAGE_SIZE)
memset(buf + fillsize, 0, PAGE_SIZE - fillsize);
if (ret == 0)
SetPageUptodate(page);
flush_dcache_page(page);
kunmap(page);
unlock_page(page);
return ret;
}
static const struct address_space_operations romfs_aops = {
.read_folio = romfs_read_folio
};
/*
* read the entries from a directory
*/
static int romfs_readdir(struct file *file, struct dir_context *ctx)
{
struct inode *i = file_inode(file);
struct romfs_inode ri;
unsigned long offset, maxoff;
int j, ino, nextfh;
char fsname[ROMFS_MAXFN]; /* XXX dynamic? */
int ret;
maxoff = romfs_maxsize(i->i_sb);
offset = ctx->pos;
if (!offset) {
offset = i->i_ino & ROMFH_MASK;
ret = romfs_dev_read(i->i_sb, offset, &ri, ROMFH_SIZE);
if (ret < 0)
goto out;
offset = be32_to_cpu(ri.spec) & ROMFH_MASK;
}
/* Not really failsafe, but we are read-only... */
for (;;) {
if (!offset || offset >= maxoff) {
offset = maxoff;
ctx->pos = offset;
goto out;
}
ctx->pos = offset;
/* Fetch inode info */
ret = romfs_dev_read(i->i_sb, offset, &ri, ROMFH_SIZE);
if (ret < 0)
goto out;
j = romfs_dev_strnlen(i->i_sb, offset + ROMFH_SIZE,
sizeof(fsname) - 1);
if (j < 0)
goto out;
ret = romfs_dev_read(i->i_sb, offset + ROMFH_SIZE, fsname, j);
if (ret < 0)
goto out;
fsname[j] = '\0';
ino = offset;
nextfh = be32_to_cpu(ri.next);
if ((nextfh & ROMFH_TYPE) == ROMFH_HRD)
ino = be32_to_cpu(ri.spec);
if (!dir_emit(ctx, fsname, j, ino,
romfs_dtype_table[nextfh & ROMFH_TYPE]))
goto out;
offset = nextfh & ROMFH_MASK;
}
out:
return 0;
}
/*
* look up an entry in a directory
*/
static struct dentry *romfs_lookup(struct inode *dir, struct dentry *dentry,
unsigned int flags)
{
unsigned long offset, maxoff;
struct inode *inode = NULL;
struct romfs_inode ri;
const char *name; /* got from dentry */
int len, ret;
offset = dir->i_ino & ROMFH_MASK;
ret = romfs_dev_read(dir->i_sb, offset, &ri, ROMFH_SIZE);
if (ret < 0)
goto error;
/* search all the file entries in the list starting from the one
* pointed to by the directory's special data */
maxoff = romfs_maxsize(dir->i_sb);
offset = be32_to_cpu(ri.spec) & ROMFH_MASK;
name = dentry->d_name.name;
len = dentry->d_name.len;
for (;;) {
if (!offset || offset >= maxoff)
break;
ret = romfs_dev_read(dir->i_sb, offset, &ri, sizeof(ri));
if (ret < 0)
goto error;
/* try to match the first 16 bytes of name */
ret = romfs_dev_strcmp(dir->i_sb, offset + ROMFH_SIZE, name,
len);
if (ret < 0)
goto error;
if (ret == 1) {
/* Hard link handling */
if ((be32_to_cpu(ri.next) & ROMFH_TYPE) == ROMFH_HRD)
offset = be32_to_cpu(ri.spec) & ROMFH_MASK;
inode = romfs_iget(dir->i_sb, offset);
break;
}
/* next entry */
offset = be32_to_cpu(ri.next) & ROMFH_MASK;
}
return d_splice_alias(inode, dentry);
error:
return ERR_PTR(ret);
}
static const struct file_operations romfs_dir_operations = {
.read = generic_read_dir,
.iterate_shared = romfs_readdir,
.llseek = generic_file_llseek,
};
static const struct inode_operations romfs_dir_inode_operations = {
.lookup = romfs_lookup,
};
/*
* get a romfs inode based on its position in the image (which doubles as the
* inode number)
*/
static struct inode *romfs_iget(struct super_block *sb, unsigned long pos)
{
struct romfs_inode_info *inode;
struct romfs_inode ri;
struct inode *i;
unsigned long nlen;
unsigned nextfh;
int ret;
umode_t mode;
/* we might have to traverse a chain of "hard link" file entries to get
* to the actual file */
for (;;) {
ret = romfs_dev_read(sb, pos, &ri, sizeof(ri));
if (ret < 0)
goto error;
/* XXX: do romfs_checksum here too (with name) */
nextfh = be32_to_cpu(ri.next);
if ((nextfh & ROMFH_TYPE) != ROMFH_HRD)
break;
pos = be32_to_cpu(ri.spec) & ROMFH_MASK;
}
/* determine the length of the filename */
nlen = romfs_dev_strnlen(sb, pos + ROMFH_SIZE, ROMFS_MAXFN);
if (IS_ERR_VALUE(nlen))
goto eio;
/* get an inode for this image position */
i = iget_locked(sb, pos);
if (!i)
return ERR_PTR(-ENOMEM);
if (!(i->i_state & I_NEW))
return i;
/* precalculate the data offset */
inode = ROMFS_I(i);
inode->i_metasize = (ROMFH_SIZE + nlen + 1 + ROMFH_PAD) & ROMFH_MASK;
inode->i_dataoffset = pos + inode->i_metasize;
set_nlink(i, 1); /* Hard to decide.. */
i->i_size = be32_to_cpu(ri.size);
i->i_mtime = i->i_atime = inode_set_ctime(i, 0, 0);
/* set up mode and ops */
mode = romfs_modemap[nextfh & ROMFH_TYPE];
switch (nextfh & ROMFH_TYPE) {
case ROMFH_DIR:
i->i_size = ROMFS_I(i)->i_metasize;
i->i_op = &romfs_dir_inode_operations;
i->i_fop = &romfs_dir_operations;
if (nextfh & ROMFH_EXEC)
mode |= S_IXUGO;
break;
case ROMFH_REG:
i->i_fop = &romfs_ro_fops;
i->i_data.a_ops = &romfs_aops;
if (nextfh & ROMFH_EXEC)
mode |= S_IXUGO;
break;
case ROMFH_SYM:
i->i_op = &page_symlink_inode_operations;
inode_nohighmem(i);
i->i_data.a_ops = &romfs_aops;
mode |= S_IRWXUGO;
break;
default:
/* depending on MBZ for sock/fifos */
nextfh = be32_to_cpu(ri.spec);
init_special_inode(i, mode, MKDEV(nextfh >> 16,
nextfh & 0xffff));
break;
}
i->i_mode = mode;
i->i_blocks = (i->i_size + 511) >> 9;
unlock_new_inode(i);
return i;
eio:
ret = -EIO;
error:
pr_err("read error for inode 0x%lx\n", pos);
return ERR_PTR(ret);
}
/*
* allocate a new inode
*/
static struct inode *romfs_alloc_inode(struct super_block *sb)
{
struct romfs_inode_info *inode;
inode = alloc_inode_sb(sb, romfs_inode_cachep, GFP_KERNEL);
return inode ? &inode->vfs_inode : NULL;
}
/*
* return a spent inode to the slab cache
*/
static void romfs_free_inode(struct inode *inode)
{
kmem_cache_free(romfs_inode_cachep, ROMFS_I(inode));
}
/*
* get filesystem statistics
*/
static int romfs_statfs(struct dentry *dentry, struct kstatfs *buf)
{
struct super_block *sb = dentry->d_sb;
u64 id = 0;
/* When calling huge_encode_dev(),
* use sb->s_bdev->bd_dev when,
* - CONFIG_ROMFS_ON_BLOCK defined
* use sb->s_dev when,
* - CONFIG_ROMFS_ON_BLOCK undefined and
* - CONFIG_ROMFS_ON_MTD defined
* leave id as 0 when,
* - CONFIG_ROMFS_ON_BLOCK undefined and
* - CONFIG_ROMFS_ON_MTD undefined
*/
if (sb->s_bdev)
id = huge_encode_dev(sb->s_bdev->bd_dev);
else if (sb->s_dev)
id = huge_encode_dev(sb->s_dev);
buf->f_type = ROMFS_MAGIC;
buf->f_namelen = ROMFS_MAXFN;
buf->f_bsize = ROMBSIZE;
buf->f_bfree = buf->f_bavail = buf->f_ffree;
buf->f_blocks =
(romfs_maxsize(dentry->d_sb) + ROMBSIZE - 1) >> ROMBSBITS;
buf->f_fsid = u64_to_fsid(id);
return 0;
}
/*
* remounting must involve read-only
*/
static int romfs_reconfigure(struct fs_context *fc)
{
sync_filesystem(fc->root->d_sb);
fc->sb_flags |= SB_RDONLY;
return 0;
}
static const struct super_operations romfs_super_ops = {
.alloc_inode = romfs_alloc_inode,
.free_inode = romfs_free_inode,
.statfs = romfs_statfs,
};
/*
* checksum check on part of a romfs filesystem
*/
static __u32 romfs_checksum(const void *data, int size)
{
const __be32 *ptr = data;
__u32 sum;
sum = 0;
size >>= 2;
while (size > 0) {
sum += be32_to_cpu(*ptr++);
size--;
}
return sum;
}
/*
* fill in the superblock
*/
static int romfs_fill_super(struct super_block *sb, struct fs_context *fc)
{
struct romfs_super_block *rsb;
struct inode *root;
unsigned long pos, img_size;
const char *storage;
size_t len;
int ret;
#ifdef CONFIG_BLOCK
if (!sb->s_mtd) {
sb_set_blocksize(sb, ROMBSIZE);
} else {
sb->s_blocksize = ROMBSIZE;
sb->s_blocksize_bits = blksize_bits(ROMBSIZE);
}
#endif
sb->s_maxbytes = 0xFFFFFFFF;
sb->s_magic = ROMFS_MAGIC;
sb->s_flags |= SB_RDONLY | SB_NOATIME;
sb->s_time_min = 0;
sb->s_time_max = 0;
sb->s_op = &romfs_super_ops;
#ifdef CONFIG_ROMFS_ON_MTD
/* Use same dev ID from the underlying mtdblock device */
if (sb->s_mtd)
sb->s_dev = MKDEV(MTD_BLOCK_MAJOR, sb->s_mtd->index);
#endif
/* read the image superblock and check it */
rsb = kmalloc(512, GFP_KERNEL);
if (!rsb)
return -ENOMEM;
sb->s_fs_info = (void *) 512;
ret = romfs_dev_read(sb, 0, rsb, 512);
if (ret < 0)
goto error_rsb;
img_size = be32_to_cpu(rsb->size);
if (sb->s_mtd && img_size > sb->s_mtd->size)
goto error_rsb_inval;
sb->s_fs_info = (void *) img_size;
if (rsb->word0 != ROMSB_WORD0 || rsb->word1 != ROMSB_WORD1 ||
img_size < ROMFH_SIZE) {
if (!(fc->sb_flags & SB_SILENT))
errorf(fc, "VFS: Can't find a romfs filesystem on dev %s.\n",
sb->s_id);
goto error_rsb_inval;
}
if (romfs_checksum(rsb, min_t(size_t, img_size, 512))) {
pr_err("bad initial checksum on dev %s.\n", sb->s_id);
goto error_rsb_inval;
}
storage = sb->s_mtd ? "MTD" : "the block layer";
len = strnlen(rsb->name, ROMFS_MAXFN);
if (!(fc->sb_flags & SB_SILENT))
pr_notice("Mounting image '%*.*s' through %s\n",
(unsigned) len, (unsigned) len, rsb->name, storage);
kfree(rsb);
rsb = NULL;
/* find the root directory */
pos = (ROMFH_SIZE + len + 1 + ROMFH_PAD) & ROMFH_MASK;
root = romfs_iget(sb, pos);
if (IS_ERR(root))
return PTR_ERR(root);
sb->s_root = d_make_root(root);
if (!sb->s_root)
return -ENOMEM;
return 0;
error_rsb_inval:
ret = -EINVAL;
error_rsb:
kfree(rsb);
return ret;
}
/*
* get a superblock for mounting
*/
static int romfs_get_tree(struct fs_context *fc)
{
int ret = -EINVAL;
#ifdef CONFIG_ROMFS_ON_MTD
ret = get_tree_mtd(fc, romfs_fill_super);
#endif
#ifdef CONFIG_ROMFS_ON_BLOCK
if (ret == -EINVAL)
ret = get_tree_bdev(fc, romfs_fill_super);
#endif
return ret;
}
static const struct fs_context_operations romfs_context_ops = {
.get_tree = romfs_get_tree,
.reconfigure = romfs_reconfigure,
};
/*
* Set up the filesystem mount context.
*/
static int romfs_init_fs_context(struct fs_context *fc)
{
fc->ops = &romfs_context_ops;
return 0;
}
/*
* destroy a romfs superblock in the appropriate manner
*/
static void romfs_kill_sb(struct super_block *sb)
{
generic_shutdown_super(sb);
#ifdef CONFIG_ROMFS_ON_MTD
if (sb->s_mtd) {
put_mtd_device(sb->s_mtd);
sb->s_mtd = NULL;
}
#endif
#ifdef CONFIG_ROMFS_ON_BLOCK
if (sb->s_bdev) {
sync_blockdev(sb->s_bdev);
blkdev_put(sb->s_bdev, sb);
}
#endif
}
static struct file_system_type romfs_fs_type = {
.owner = THIS_MODULE,
.name = "romfs",
.init_fs_context = romfs_init_fs_context,
.kill_sb = romfs_kill_sb,
.fs_flags = FS_REQUIRES_DEV,
};
MODULE_ALIAS_FS("romfs");
/*
* inode storage initialiser
*/
static void romfs_i_init_once(void *_inode)
{
struct romfs_inode_info *inode = _inode;
inode_init_once(&inode->vfs_inode);
}
/*
* romfs module initialisation
*/
static int __init init_romfs_fs(void)
{
int ret;
pr_info("ROMFS MTD (C) 2007 Red Hat, Inc.\n");
romfs_inode_cachep =
kmem_cache_create("romfs_i",
sizeof(struct romfs_inode_info), 0,
SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD |
SLAB_ACCOUNT, romfs_i_init_once);
if (!romfs_inode_cachep) {
pr_err("Failed to initialise inode cache\n");
return -ENOMEM;
}
ret = register_filesystem(&romfs_fs_type);
if (ret) {
pr_err("Failed to register filesystem\n");
goto error_register;
}
return 0;
error_register:
kmem_cache_destroy(romfs_inode_cachep);
return ret;
}
/*
* romfs module removal
*/
static void __exit exit_romfs_fs(void)
{
unregister_filesystem(&romfs_fs_type);
/*
* Make sure all delayed rcu free inodes are flushed before we
* destroy cache.
*/
rcu_barrier();
kmem_cache_destroy(romfs_inode_cachep);
}
module_init(init_romfs_fs);
module_exit(exit_romfs_fs);
MODULE_DESCRIPTION("Direct-MTD Capable RomFS");
MODULE_AUTHOR("Red Hat, Inc.");
MODULE_LICENSE("GPL"); /* Actually dual-licensed, but it doesn't matter for */