linux-stable/fs/erofs/xattr.c
Jingbo Xu 3acea5fc33 erofs: avoid hardcoded blocksize for subpage block support
As the first step of converting hardcoded blocksize to that specified in
on-disk superblock, convert all call sites of hardcoded blocksize to
sb->s_blocksize except for:

1) use sbi->blkszbits instead of sb->s_blocksize in
erofs_superblock_csum_verify() since sb->s_blocksize has not been
updated with the on-disk blocksize yet when the function is called.

2) use inode->i_blkbits instead of sb->s_blocksize in erofs_bread(),
since the inode operated on may be an anonymous inode in fscache mode.
Currently the anonymous inode is allocated from an anonymous mount
maintained in erofs, while in the near future we may allocate anonymous
inodes from a generic API directly and thus have no access to the
anonymous inode's i_sb.  Thus we keep the block size in i_blkbits for
anonymous inodes in fscache mode.

Be noted that this patch only gets rid of the hardcoded blocksize, in
preparation for actually setting the on-disk block size in the following
patch.  The hard limit of constraining the block size to PAGE_SIZE still
exists until the next patch.

Signed-off-by: Jingbo Xu <jefflexu@linux.alibaba.com>
Reviewed-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Reviewed-by: Yue Hu <huyue2@coolpad.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Link: https://lore.kernel.org/r/20230313135309.75269-2-jefflexu@linux.alibaba.com
[ Gao Xiang: fold a patch to fix incorrect truncated offsets. ]
Link: https://lore.kernel.org/r/20230413035734.15457-1-zhujia.zj@bytedance.com
Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
2023-04-17 01:15:44 +08:00

652 lines
16 KiB
C

// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright (C) 2017-2018 HUAWEI, Inc.
* https://www.huawei.com/
* Copyright (C) 2021-2022, Alibaba Cloud
*/
#include <linux/security.h>
#include "xattr.h"
struct xattr_iter {
struct super_block *sb;
struct erofs_buf buf;
void *kaddr;
erofs_blk_t blkaddr;
unsigned int ofs;
};
static int init_inode_xattrs(struct inode *inode)
{
struct erofs_inode *const vi = EROFS_I(inode);
struct xattr_iter it;
unsigned int i;
struct erofs_xattr_ibody_header *ih;
struct super_block *sb = inode->i_sb;
int ret = 0;
/* the most case is that xattrs of this inode are initialized. */
if (test_bit(EROFS_I_EA_INITED_BIT, &vi->flags)) {
/*
* paired with smp_mb() at the end of the function to ensure
* fields will only be observed after the bit is set.
*/
smp_mb();
return 0;
}
if (wait_on_bit_lock(&vi->flags, EROFS_I_BL_XATTR_BIT, TASK_KILLABLE))
return -ERESTARTSYS;
/* someone has initialized xattrs for us? */
if (test_bit(EROFS_I_EA_INITED_BIT, &vi->flags))
goto out_unlock;
/*
* bypass all xattr operations if ->xattr_isize is not greater than
* sizeof(struct erofs_xattr_ibody_header), in detail:
* 1) it is not enough to contain erofs_xattr_ibody_header then
* ->xattr_isize should be 0 (it means no xattr);
* 2) it is just to contain erofs_xattr_ibody_header, which is on-disk
* undefined right now (maybe use later with some new sb feature).
*/
if (vi->xattr_isize == sizeof(struct erofs_xattr_ibody_header)) {
erofs_err(sb,
"xattr_isize %d of nid %llu is not supported yet",
vi->xattr_isize, vi->nid);
ret = -EOPNOTSUPP;
goto out_unlock;
} else if (vi->xattr_isize < sizeof(struct erofs_xattr_ibody_header)) {
if (vi->xattr_isize) {
erofs_err(sb, "bogus xattr ibody @ nid %llu", vi->nid);
DBG_BUGON(1);
ret = -EFSCORRUPTED;
goto out_unlock; /* xattr ondisk layout error */
}
ret = -ENOATTR;
goto out_unlock;
}
it.buf = __EROFS_BUF_INITIALIZER;
it.blkaddr = erofs_blknr(sb, erofs_iloc(inode) + vi->inode_isize);
it.ofs = erofs_blkoff(sb, erofs_iloc(inode) + vi->inode_isize);
/* read in shared xattr array (non-atomic, see kmalloc below) */
it.kaddr = erofs_read_metabuf(&it.buf, sb, it.blkaddr, EROFS_KMAP);
if (IS_ERR(it.kaddr)) {
ret = PTR_ERR(it.kaddr);
goto out_unlock;
}
ih = (struct erofs_xattr_ibody_header *)(it.kaddr + it.ofs);
vi->xattr_shared_count = ih->h_shared_count;
vi->xattr_shared_xattrs = kmalloc_array(vi->xattr_shared_count,
sizeof(uint), GFP_KERNEL);
if (!vi->xattr_shared_xattrs) {
erofs_put_metabuf(&it.buf);
ret = -ENOMEM;
goto out_unlock;
}
/* let's skip ibody header */
it.ofs += sizeof(struct erofs_xattr_ibody_header);
for (i = 0; i < vi->xattr_shared_count; ++i) {
if (it.ofs >= sb->s_blocksize) {
/* cannot be unaligned */
DBG_BUGON(it.ofs != sb->s_blocksize);
it.kaddr = erofs_read_metabuf(&it.buf, sb, ++it.blkaddr,
EROFS_KMAP);
if (IS_ERR(it.kaddr)) {
kfree(vi->xattr_shared_xattrs);
vi->xattr_shared_xattrs = NULL;
ret = PTR_ERR(it.kaddr);
goto out_unlock;
}
it.ofs = 0;
}
vi->xattr_shared_xattrs[i] =
le32_to_cpu(*(__le32 *)(it.kaddr + it.ofs));
it.ofs += sizeof(__le32);
}
erofs_put_metabuf(&it.buf);
/* paired with smp_mb() at the beginning of the function. */
smp_mb();
set_bit(EROFS_I_EA_INITED_BIT, &vi->flags);
out_unlock:
clear_and_wake_up_bit(EROFS_I_BL_XATTR_BIT, &vi->flags);
return ret;
}
/*
* the general idea for these return values is
* if 0 is returned, go on processing the current xattr;
* 1 (> 0) is returned, skip this round to process the next xattr;
* -err (< 0) is returned, an error (maybe ENOXATTR) occurred
* and need to be handled
*/
struct xattr_iter_handlers {
int (*entry)(struct xattr_iter *_it, struct erofs_xattr_entry *entry);
int (*name)(struct xattr_iter *_it, unsigned int processed, char *buf,
unsigned int len);
int (*alloc_buffer)(struct xattr_iter *_it, unsigned int value_sz);
void (*value)(struct xattr_iter *_it, unsigned int processed, char *buf,
unsigned int len);
};
static inline int xattr_iter_fixup(struct xattr_iter *it)
{
if (it->ofs < it->sb->s_blocksize)
return 0;
it->blkaddr += erofs_blknr(it->sb, it->ofs);
it->kaddr = erofs_read_metabuf(&it->buf, it->sb, it->blkaddr,
EROFS_KMAP);
if (IS_ERR(it->kaddr))
return PTR_ERR(it->kaddr);
it->ofs = erofs_blkoff(it->sb, it->ofs);
return 0;
}
static int inline_xattr_iter_begin(struct xattr_iter *it,
struct inode *inode)
{
struct erofs_inode *const vi = EROFS_I(inode);
unsigned int xattr_header_sz, inline_xattr_ofs;
xattr_header_sz = inlinexattr_header_size(inode);
if (xattr_header_sz >= vi->xattr_isize) {
DBG_BUGON(xattr_header_sz > vi->xattr_isize);
return -ENOATTR;
}
inline_xattr_ofs = vi->inode_isize + xattr_header_sz;
it->blkaddr = erofs_blknr(it->sb, erofs_iloc(inode) + inline_xattr_ofs);
it->ofs = erofs_blkoff(it->sb, erofs_iloc(inode) + inline_xattr_ofs);
it->kaddr = erofs_read_metabuf(&it->buf, inode->i_sb, it->blkaddr,
EROFS_KMAP);
if (IS_ERR(it->kaddr))
return PTR_ERR(it->kaddr);
return vi->xattr_isize - xattr_header_sz;
}
/*
* Regardless of success or failure, `xattr_foreach' will end up with
* `ofs' pointing to the next xattr item rather than an arbitrary position.
*/
static int xattr_foreach(struct xattr_iter *it,
const struct xattr_iter_handlers *op,
unsigned int *tlimit)
{
struct erofs_xattr_entry entry;
unsigned int value_sz, processed, slice;
int err;
/* 0. fixup blkaddr, ofs, ipage */
err = xattr_iter_fixup(it);
if (err)
return err;
/*
* 1. read xattr entry to the memory,
* since we do EROFS_XATTR_ALIGN
* therefore entry should be in the page
*/
entry = *(struct erofs_xattr_entry *)(it->kaddr + it->ofs);
if (tlimit) {
unsigned int entry_sz = erofs_xattr_entry_size(&entry);
/* xattr on-disk corruption: xattr entry beyond xattr_isize */
if (*tlimit < entry_sz) {
DBG_BUGON(1);
return -EFSCORRUPTED;
}
*tlimit -= entry_sz;
}
it->ofs += sizeof(struct erofs_xattr_entry);
value_sz = le16_to_cpu(entry.e_value_size);
/* handle entry */
err = op->entry(it, &entry);
if (err) {
it->ofs += entry.e_name_len + value_sz;
goto out;
}
/* 2. handle xattr name (ofs will finally be at the end of name) */
processed = 0;
while (processed < entry.e_name_len) {
if (it->ofs >= it->sb->s_blocksize) {
DBG_BUGON(it->ofs > it->sb->s_blocksize);
err = xattr_iter_fixup(it);
if (err)
goto out;
it->ofs = 0;
}
slice = min_t(unsigned int, it->sb->s_blocksize - it->ofs,
entry.e_name_len - processed);
/* handle name */
err = op->name(it, processed, it->kaddr + it->ofs, slice);
if (err) {
it->ofs += entry.e_name_len - processed + value_sz;
goto out;
}
it->ofs += slice;
processed += slice;
}
/* 3. handle xattr value */
processed = 0;
if (op->alloc_buffer) {
err = op->alloc_buffer(it, value_sz);
if (err) {
it->ofs += value_sz;
goto out;
}
}
while (processed < value_sz) {
if (it->ofs >= it->sb->s_blocksize) {
DBG_BUGON(it->ofs > it->sb->s_blocksize);
err = xattr_iter_fixup(it);
if (err)
goto out;
it->ofs = 0;
}
slice = min_t(unsigned int, it->sb->s_blocksize - it->ofs,
value_sz - processed);
op->value(it, processed, it->kaddr + it->ofs, slice);
it->ofs += slice;
processed += slice;
}
out:
/* xattrs should be 4-byte aligned (on-disk constraint) */
it->ofs = EROFS_XATTR_ALIGN(it->ofs);
return err < 0 ? err : 0;
}
struct getxattr_iter {
struct xattr_iter it;
char *buffer;
int buffer_size, index;
struct qstr name;
};
static int xattr_entrymatch(struct xattr_iter *_it,
struct erofs_xattr_entry *entry)
{
struct getxattr_iter *it = container_of(_it, struct getxattr_iter, it);
return (it->index != entry->e_name_index ||
it->name.len != entry->e_name_len) ? -ENOATTR : 0;
}
static int xattr_namematch(struct xattr_iter *_it,
unsigned int processed, char *buf, unsigned int len)
{
struct getxattr_iter *it = container_of(_it, struct getxattr_iter, it);
return memcmp(buf, it->name.name + processed, len) ? -ENOATTR : 0;
}
static int xattr_checkbuffer(struct xattr_iter *_it,
unsigned int value_sz)
{
struct getxattr_iter *it = container_of(_it, struct getxattr_iter, it);
int err = it->buffer_size < value_sz ? -ERANGE : 0;
it->buffer_size = value_sz;
return !it->buffer ? 1 : err;
}
static void xattr_copyvalue(struct xattr_iter *_it,
unsigned int processed,
char *buf, unsigned int len)
{
struct getxattr_iter *it = container_of(_it, struct getxattr_iter, it);
memcpy(it->buffer + processed, buf, len);
}
static const struct xattr_iter_handlers find_xattr_handlers = {
.entry = xattr_entrymatch,
.name = xattr_namematch,
.alloc_buffer = xattr_checkbuffer,
.value = xattr_copyvalue
};
static int inline_getxattr(struct inode *inode, struct getxattr_iter *it)
{
int ret;
unsigned int remaining;
ret = inline_xattr_iter_begin(&it->it, inode);
if (ret < 0)
return ret;
remaining = ret;
while (remaining) {
ret = xattr_foreach(&it->it, &find_xattr_handlers, &remaining);
if (ret != -ENOATTR)
break;
}
return ret ? ret : it->buffer_size;
}
static int shared_getxattr(struct inode *inode, struct getxattr_iter *it)
{
struct erofs_inode *const vi = EROFS_I(inode);
struct super_block *const sb = inode->i_sb;
unsigned int i;
int ret = -ENOATTR;
for (i = 0; i < vi->xattr_shared_count; ++i) {
erofs_blk_t blkaddr =
xattrblock_addr(sb, vi->xattr_shared_xattrs[i]);
it->it.ofs = xattrblock_offset(sb, vi->xattr_shared_xattrs[i]);
it->it.kaddr = erofs_read_metabuf(&it->it.buf, sb, blkaddr,
EROFS_KMAP);
if (IS_ERR(it->it.kaddr))
return PTR_ERR(it->it.kaddr);
it->it.blkaddr = blkaddr;
ret = xattr_foreach(&it->it, &find_xattr_handlers, NULL);
if (ret != -ENOATTR)
break;
}
return ret ? ret : it->buffer_size;
}
static bool erofs_xattr_user_list(struct dentry *dentry)
{
return test_opt(&EROFS_SB(dentry->d_sb)->opt, XATTR_USER);
}
static bool erofs_xattr_trusted_list(struct dentry *dentry)
{
return capable(CAP_SYS_ADMIN);
}
int erofs_getxattr(struct inode *inode, int index,
const char *name,
void *buffer, size_t buffer_size)
{
int ret;
struct getxattr_iter it;
if (!name)
return -EINVAL;
ret = init_inode_xattrs(inode);
if (ret)
return ret;
it.index = index;
it.name.len = strlen(name);
if (it.name.len > EROFS_NAME_LEN)
return -ERANGE;
it.it.buf = __EROFS_BUF_INITIALIZER;
it.name.name = name;
it.buffer = buffer;
it.buffer_size = buffer_size;
it.it.sb = inode->i_sb;
ret = inline_getxattr(inode, &it);
if (ret == -ENOATTR)
ret = shared_getxattr(inode, &it);
erofs_put_metabuf(&it.it.buf);
return ret;
}
static int erofs_xattr_generic_get(const struct xattr_handler *handler,
struct dentry *unused, struct inode *inode,
const char *name, void *buffer, size_t size)
{
struct erofs_sb_info *const sbi = EROFS_I_SB(inode);
switch (handler->flags) {
case EROFS_XATTR_INDEX_USER:
if (!test_opt(&sbi->opt, XATTR_USER))
return -EOPNOTSUPP;
break;
case EROFS_XATTR_INDEX_TRUSTED:
break;
case EROFS_XATTR_INDEX_SECURITY:
break;
default:
return -EINVAL;
}
return erofs_getxattr(inode, handler->flags, name, buffer, size);
}
const struct xattr_handler erofs_xattr_user_handler = {
.prefix = XATTR_USER_PREFIX,
.flags = EROFS_XATTR_INDEX_USER,
.list = erofs_xattr_user_list,
.get = erofs_xattr_generic_get,
};
const struct xattr_handler erofs_xattr_trusted_handler = {
.prefix = XATTR_TRUSTED_PREFIX,
.flags = EROFS_XATTR_INDEX_TRUSTED,
.list = erofs_xattr_trusted_list,
.get = erofs_xattr_generic_get,
};
#ifdef CONFIG_EROFS_FS_SECURITY
const struct xattr_handler __maybe_unused erofs_xattr_security_handler = {
.prefix = XATTR_SECURITY_PREFIX,
.flags = EROFS_XATTR_INDEX_SECURITY,
.get = erofs_xattr_generic_get,
};
#endif
const struct xattr_handler *erofs_xattr_handlers[] = {
&erofs_xattr_user_handler,
#ifdef CONFIG_EROFS_FS_POSIX_ACL
&posix_acl_access_xattr_handler,
&posix_acl_default_xattr_handler,
#endif
&erofs_xattr_trusted_handler,
#ifdef CONFIG_EROFS_FS_SECURITY
&erofs_xattr_security_handler,
#endif
NULL,
};
struct listxattr_iter {
struct xattr_iter it;
struct dentry *dentry;
char *buffer;
int buffer_size, buffer_ofs;
};
static int xattr_entrylist(struct xattr_iter *_it,
struct erofs_xattr_entry *entry)
{
struct listxattr_iter *it =
container_of(_it, struct listxattr_iter, it);
unsigned int prefix_len;
const char *prefix;
const struct xattr_handler *h =
erofs_xattr_handler(entry->e_name_index);
if (!h || (h->list && !h->list(it->dentry)))
return 1;
prefix = xattr_prefix(h);
prefix_len = strlen(prefix);
if (!it->buffer) {
it->buffer_ofs += prefix_len + entry->e_name_len + 1;
return 1;
}
if (it->buffer_ofs + prefix_len
+ entry->e_name_len + 1 > it->buffer_size)
return -ERANGE;
memcpy(it->buffer + it->buffer_ofs, prefix, prefix_len);
it->buffer_ofs += prefix_len;
return 0;
}
static int xattr_namelist(struct xattr_iter *_it,
unsigned int processed, char *buf, unsigned int len)
{
struct listxattr_iter *it =
container_of(_it, struct listxattr_iter, it);
memcpy(it->buffer + it->buffer_ofs, buf, len);
it->buffer_ofs += len;
return 0;
}
static int xattr_skipvalue(struct xattr_iter *_it,
unsigned int value_sz)
{
struct listxattr_iter *it =
container_of(_it, struct listxattr_iter, it);
it->buffer[it->buffer_ofs++] = '\0';
return 1;
}
static const struct xattr_iter_handlers list_xattr_handlers = {
.entry = xattr_entrylist,
.name = xattr_namelist,
.alloc_buffer = xattr_skipvalue,
.value = NULL
};
static int inline_listxattr(struct listxattr_iter *it)
{
int ret;
unsigned int remaining;
ret = inline_xattr_iter_begin(&it->it, d_inode(it->dentry));
if (ret < 0)
return ret;
remaining = ret;
while (remaining) {
ret = xattr_foreach(&it->it, &list_xattr_handlers, &remaining);
if (ret)
break;
}
return ret ? ret : it->buffer_ofs;
}
static int shared_listxattr(struct listxattr_iter *it)
{
struct inode *const inode = d_inode(it->dentry);
struct erofs_inode *const vi = EROFS_I(inode);
struct super_block *const sb = inode->i_sb;
unsigned int i;
int ret = 0;
for (i = 0; i < vi->xattr_shared_count; ++i) {
erofs_blk_t blkaddr =
xattrblock_addr(sb, vi->xattr_shared_xattrs[i]);
it->it.ofs = xattrblock_offset(sb, vi->xattr_shared_xattrs[i]);
it->it.kaddr = erofs_read_metabuf(&it->it.buf, sb, blkaddr,
EROFS_KMAP);
if (IS_ERR(it->it.kaddr))
return PTR_ERR(it->it.kaddr);
it->it.blkaddr = blkaddr;
ret = xattr_foreach(&it->it, &list_xattr_handlers, NULL);
if (ret)
break;
}
return ret ? ret : it->buffer_ofs;
}
ssize_t erofs_listxattr(struct dentry *dentry,
char *buffer, size_t buffer_size)
{
int ret;
struct listxattr_iter it;
ret = init_inode_xattrs(d_inode(dentry));
if (ret == -ENOATTR)
return 0;
if (ret)
return ret;
it.it.buf = __EROFS_BUF_INITIALIZER;
it.dentry = dentry;
it.buffer = buffer;
it.buffer_size = buffer_size;
it.buffer_ofs = 0;
it.it.sb = dentry->d_sb;
ret = inline_listxattr(&it);
if (ret >= 0 || ret == -ENOATTR)
ret = shared_listxattr(&it);
erofs_put_metabuf(&it.it.buf);
return ret;
}
#ifdef CONFIG_EROFS_FS_POSIX_ACL
struct posix_acl *erofs_get_acl(struct inode *inode, int type, bool rcu)
{
struct posix_acl *acl;
int prefix, rc;
char *value = NULL;
if (rcu)
return ERR_PTR(-ECHILD);
switch (type) {
case ACL_TYPE_ACCESS:
prefix = EROFS_XATTR_INDEX_POSIX_ACL_ACCESS;
break;
case ACL_TYPE_DEFAULT:
prefix = EROFS_XATTR_INDEX_POSIX_ACL_DEFAULT;
break;
default:
return ERR_PTR(-EINVAL);
}
rc = erofs_getxattr(inode, prefix, "", NULL, 0);
if (rc > 0) {
value = kmalloc(rc, GFP_KERNEL);
if (!value)
return ERR_PTR(-ENOMEM);
rc = erofs_getxattr(inode, prefix, "", value, rc);
}
if (rc == -ENOATTR)
acl = NULL;
else if (rc < 0)
acl = ERR_PTR(rc);
else
acl = posix_acl_from_xattr(&init_user_ns, value, rc);
kfree(value);
return acl;
}
#endif