linux_dsm_epyc7002/fs/minix/dir.c
Nick Piggin 6fe6900e1e mm: make read_cache_page synchronous
Ensure pages are uptodate after returning from read_cache_page, which allows
us to cut out most of the filesystem-internal PageUptodate calls.

I didn't have a great look down the call chains, but this appears to fixes 7
possible use-before uptodate in hfs, 2 in hfsplus, 1 in jfs, a few in
ecryptfs, 1 in jffs2, and a possible cleared data overwritten with readpage in
block2mtd.  All depending on whether the filler is async and/or can return
with a !uptodate page.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-07 12:12:51 -07:00

468 lines
11 KiB
C

/*
* linux/fs/minix/dir.c
*
* Copyright (C) 1991, 1992 Linus Torvalds
*
* minix directory handling functions
*
* Updated to filesystem version 3 by Daniel Aragones
*/
#include "minix.h"
#include <linux/highmem.h>
#include <linux/smp_lock.h>
typedef struct minix_dir_entry minix_dirent;
typedef struct minix3_dir_entry minix3_dirent;
static int minix_readdir(struct file *, void *, filldir_t);
const struct file_operations minix_dir_operations = {
.read = generic_read_dir,
.readdir = minix_readdir,
.fsync = minix_sync_file,
};
static inline void dir_put_page(struct page *page)
{
kunmap(page);
page_cache_release(page);
}
/*
* Return the offset into page `page_nr' of the last valid
* byte in that page, plus one.
*/
static unsigned
minix_last_byte(struct inode *inode, unsigned long page_nr)
{
unsigned last_byte = PAGE_CACHE_SIZE;
if (page_nr == (inode->i_size >> PAGE_CACHE_SHIFT))
last_byte = inode->i_size & (PAGE_CACHE_SIZE - 1);
return last_byte;
}
static inline unsigned long dir_pages(struct inode *inode)
{
return (inode->i_size+PAGE_CACHE_SIZE-1)>>PAGE_CACHE_SHIFT;
}
static int dir_commit_chunk(struct page *page, unsigned from, unsigned to)
{
struct inode *dir = (struct inode *)page->mapping->host;
int err = 0;
page->mapping->a_ops->commit_write(NULL, page, from, to);
if (IS_DIRSYNC(dir))
err = write_one_page(page, 1);
else
unlock_page(page);
return err;
}
static struct page * dir_get_page(struct inode *dir, unsigned long n)
{
struct address_space *mapping = dir->i_mapping;
struct page *page = read_mapping_page(mapping, n, NULL);
if (!IS_ERR(page)) {
kmap(page);
if (!PageUptodate(page))
goto fail;
}
return page;
fail:
dir_put_page(page);
return ERR_PTR(-EIO);
}
static inline void *minix_next_entry(void *de, struct minix_sb_info *sbi)
{
return (void*)((char*)de + sbi->s_dirsize);
}
static int minix_readdir(struct file * filp, void * dirent, filldir_t filldir)
{
unsigned long pos = filp->f_pos;
struct inode *inode = filp->f_path.dentry->d_inode;
struct super_block *sb = inode->i_sb;
unsigned offset = pos & ~PAGE_CACHE_MASK;
unsigned long n = pos >> PAGE_CACHE_SHIFT;
unsigned long npages = dir_pages(inode);
struct minix_sb_info *sbi = minix_sb(sb);
unsigned chunk_size = sbi->s_dirsize;
char *name;
__u32 inumber;
lock_kernel();
pos = (pos + chunk_size-1) & ~(chunk_size-1);
if (pos >= inode->i_size)
goto done;
for ( ; n < npages; n++, offset = 0) {
char *p, *kaddr, *limit;
struct page *page = dir_get_page(inode, n);
if (IS_ERR(page))
continue;
kaddr = (char *)page_address(page);
p = kaddr+offset;
limit = kaddr + minix_last_byte(inode, n) - chunk_size;
for ( ; p <= limit; p = minix_next_entry(p, sbi)) {
if (sbi->s_version == MINIX_V3) {
minix3_dirent *de3 = (minix3_dirent *)p;
name = de3->name;
inumber = de3->inode;
} else {
minix_dirent *de = (minix_dirent *)p;
name = de->name;
inumber = de->inode;
}
if (inumber) {
int over;
unsigned l = strnlen(name, sbi->s_namelen);
offset = p - kaddr;
over = filldir(dirent, name, l,
(n << PAGE_CACHE_SHIFT) | offset,
inumber, DT_UNKNOWN);
if (over) {
dir_put_page(page);
goto done;
}
}
}
dir_put_page(page);
}
done:
filp->f_pos = (n << PAGE_CACHE_SHIFT) | offset;
unlock_kernel();
return 0;
}
static inline int namecompare(int len, int maxlen,
const char * name, const char * buffer)
{
if (len < maxlen && buffer[len])
return 0;
return !memcmp(name, buffer, len);
}
/*
* minix_find_entry()
*
* finds an entry in the specified directory with the wanted name. It
* returns the cache buffer in which the entry was found, and the entry
* itself (as a parameter - res_dir). It does NOT read the inode of the
* entry - you'll have to do that yourself if you want to.
*/
minix_dirent *minix_find_entry(struct dentry *dentry, struct page **res_page)
{
const char * name = dentry->d_name.name;
int namelen = dentry->d_name.len;
struct inode * dir = dentry->d_parent->d_inode;
struct super_block * sb = dir->i_sb;
struct minix_sb_info * sbi = minix_sb(sb);
unsigned long n;
unsigned long npages = dir_pages(dir);
struct page *page = NULL;
char *p;
char *namx;
__u32 inumber;
*res_page = NULL;
for (n = 0; n < npages; n++) {
char *kaddr, *limit;
page = dir_get_page(dir, n);
if (IS_ERR(page))
continue;
kaddr = (char*)page_address(page);
limit = kaddr + minix_last_byte(dir, n) - sbi->s_dirsize;
for (p = kaddr; p <= limit; p = minix_next_entry(p, sbi)) {
if (sbi->s_version == MINIX_V3) {
minix3_dirent *de3 = (minix3_dirent *)p;
namx = de3->name;
inumber = de3->inode;
} else {
minix_dirent *de = (minix_dirent *)p;
namx = de->name;
inumber = de->inode;
}
if (!inumber)
continue;
if (namecompare(namelen, sbi->s_namelen, name, namx))
goto found;
}
dir_put_page(page);
}
return NULL;
found:
*res_page = page;
return (minix_dirent *)p;
}
int minix_add_link(struct dentry *dentry, struct inode *inode)
{
struct inode *dir = dentry->d_parent->d_inode;
const char * name = dentry->d_name.name;
int namelen = dentry->d_name.len;
struct super_block * sb = dir->i_sb;
struct minix_sb_info * sbi = minix_sb(sb);
struct page *page = NULL;
unsigned long npages = dir_pages(dir);
unsigned long n;
char *kaddr, *p;
minix_dirent *de;
minix3_dirent *de3;
unsigned from, to;
int err;
char *namx = NULL;
__u32 inumber;
/*
* We take care of directory expansion in the same loop
* This code plays outside i_size, so it locks the page
* to protect that region.
*/
for (n = 0; n <= npages; n++) {
char *limit, *dir_end;
page = dir_get_page(dir, n);
err = PTR_ERR(page);
if (IS_ERR(page))
goto out;
lock_page(page);
kaddr = (char*)page_address(page);
dir_end = kaddr + minix_last_byte(dir, n);
limit = kaddr + PAGE_CACHE_SIZE - sbi->s_dirsize;
for (p = kaddr; p <= limit; p = minix_next_entry(p, sbi)) {
de = (minix_dirent *)p;
de3 = (minix3_dirent *)p;
if (sbi->s_version == MINIX_V3) {
namx = de3->name;
inumber = de3->inode;
} else {
namx = de->name;
inumber = de->inode;
}
if (p == dir_end) {
/* We hit i_size */
if (sbi->s_version == MINIX_V3)
de3->inode = 0;
else
de->inode = 0;
goto got_it;
}
if (!inumber)
goto got_it;
err = -EEXIST;
if (namecompare(namelen, sbi->s_namelen, name, namx))
goto out_unlock;
}
unlock_page(page);
dir_put_page(page);
}
BUG();
return -EINVAL;
got_it:
from = p - (char*)page_address(page);
to = from + sbi->s_dirsize;
err = page->mapping->a_ops->prepare_write(NULL, page, from, to);
if (err)
goto out_unlock;
memcpy (namx, name, namelen);
if (sbi->s_version == MINIX_V3) {
memset (namx + namelen, 0, sbi->s_dirsize - namelen - 4);
de3->inode = inode->i_ino;
} else {
memset (namx + namelen, 0, sbi->s_dirsize - namelen - 2);
de->inode = inode->i_ino;
}
err = dir_commit_chunk(page, from, to);
dir->i_mtime = dir->i_ctime = CURRENT_TIME_SEC;
mark_inode_dirty(dir);
out_put:
dir_put_page(page);
out:
return err;
out_unlock:
unlock_page(page);
goto out_put;
}
int minix_delete_entry(struct minix_dir_entry *de, struct page *page)
{
struct address_space *mapping = page->mapping;
struct inode *inode = (struct inode*)mapping->host;
char *kaddr = page_address(page);
unsigned from = (char*)de - kaddr;
unsigned to = from + minix_sb(inode->i_sb)->s_dirsize;
int err;
lock_page(page);
err = mapping->a_ops->prepare_write(NULL, page, from, to);
if (err == 0) {
de->inode = 0;
err = dir_commit_chunk(page, from, to);
} else {
unlock_page(page);
}
dir_put_page(page);
inode->i_ctime = inode->i_mtime = CURRENT_TIME_SEC;
mark_inode_dirty(inode);
return err;
}
int minix_make_empty(struct inode *inode, struct inode *dir)
{
struct address_space *mapping = inode->i_mapping;
struct page *page = grab_cache_page(mapping, 0);
struct minix_sb_info *sbi = minix_sb(inode->i_sb);
char *kaddr;
int err;
if (!page)
return -ENOMEM;
err = mapping->a_ops->prepare_write(NULL, page, 0, 2 * sbi->s_dirsize);
if (err) {
unlock_page(page);
goto fail;
}
kaddr = kmap_atomic(page, KM_USER0);
memset(kaddr, 0, PAGE_CACHE_SIZE);
if (sbi->s_version == MINIX_V3) {
minix3_dirent *de3 = (minix3_dirent *)kaddr;
de3->inode = inode->i_ino;
strcpy(de3->name, ".");
de3 = minix_next_entry(de3, sbi);
de3->inode = dir->i_ino;
strcpy(de3->name, "..");
} else {
minix_dirent *de = (minix_dirent *)kaddr;
de->inode = inode->i_ino;
strcpy(de->name, ".");
de = minix_next_entry(de, sbi);
de->inode = dir->i_ino;
strcpy(de->name, "..");
}
kunmap_atomic(kaddr, KM_USER0);
err = dir_commit_chunk(page, 0, 2 * sbi->s_dirsize);
fail:
page_cache_release(page);
return err;
}
/*
* routine to check that the specified directory is empty (for rmdir)
*/
int minix_empty_dir(struct inode * inode)
{
struct page *page = NULL;
unsigned long i, npages = dir_pages(inode);
struct minix_sb_info *sbi = minix_sb(inode->i_sb);
char *name;
__u32 inumber;
for (i = 0; i < npages; i++) {
char *p, *kaddr, *limit;
page = dir_get_page(inode, i);
if (IS_ERR(page))
continue;
kaddr = (char *)page_address(page);
limit = kaddr + minix_last_byte(inode, i) - sbi->s_dirsize;
for (p = kaddr; p <= limit; p = minix_next_entry(p, sbi)) {
if (sbi->s_version == MINIX_V3) {
minix3_dirent *de3 = (minix3_dirent *)p;
name = de3->name;
inumber = de3->inode;
} else {
minix_dirent *de = (minix_dirent *)p;
name = de->name;
inumber = de->inode;
}
if (inumber != 0) {
/* check for . and .. */
if (name[0] != '.')
goto not_empty;
if (!name[1]) {
if (inumber != inode->i_ino)
goto not_empty;
} else if (name[1] != '.')
goto not_empty;
else if (name[2])
goto not_empty;
}
}
dir_put_page(page);
}
return 1;
not_empty:
dir_put_page(page);
return 0;
}
/* Releases the page */
void minix_set_link(struct minix_dir_entry *de, struct page *page,
struct inode *inode)
{
struct inode *dir = (struct inode*)page->mapping->host;
struct minix_sb_info *sbi = minix_sb(dir->i_sb);
unsigned from = (char *)de-(char*)page_address(page);
unsigned to = from + sbi->s_dirsize;
int err;
lock_page(page);
err = page->mapping->a_ops->prepare_write(NULL, page, from, to);
if (err == 0) {
de->inode = inode->i_ino;
err = dir_commit_chunk(page, from, to);
} else {
unlock_page(page);
}
dir_put_page(page);
dir->i_mtime = dir->i_ctime = CURRENT_TIME_SEC;
mark_inode_dirty(dir);
}
struct minix_dir_entry * minix_dotdot (struct inode *dir, struct page **p)
{
struct page *page = dir_get_page(dir, 0);
struct minix_sb_info *sbi = minix_sb(dir->i_sb);
struct minix_dir_entry *de = NULL;
if (!IS_ERR(page)) {
de = minix_next_entry(page_address(page), sbi);
*p = page;
}
return de;
}
ino_t minix_inode_by_name(struct dentry *dentry)
{
struct page *page;
struct minix_dir_entry *de = minix_find_entry(dentry, &page);
ino_t res = 0;
if (de) {
res = de->inode;
dir_put_page(page);
}
return res;
}