q3k/linux
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
linux/fs/isofs/rock.c
Arnd Bergmann 4f819a7899 BKL: Remove BKL from isofs 
As in other file systems, we can replace the big kernel lock
with a private mutex in isofs. This means we can now access
multiple file systems concurrently, but it also means that
we serialize readdir and lookup across sleeping operations
which previously released the big kernel lock. This should
not matter though, as these operations are in practice
serialized through the hardware access.

The isofs_get_blocks functions now does not take any lock
any more, it used to recursively get the BKL. After looking
at the code for hours, I convinced myself that it was never
needed here anyway, because it only reads constant fields
of the inode and writes to a buffer head array that is
at this time only visible to the caller.

The get_sb and fill_super operations do not need the locking
at all because they operate on a file system that is either
about to be created or to be destroyed but in either case
is not visible to other threads.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
2010-10-04 21:10:45 +02:00

778 lines
18 KiB
C
Raw Blame History

/*
* linux/fs/isofs/rock.c
*
* (C) 1992, 1993 Eric Youngdale
*
* Rock Ridge Extensions to iso9660
*/
#include <linux/slab.h>
#include <linux/pagemap.h>
#include "isofs.h"
#include "rock.h"
/*
* These functions are designed to read the system areas of a directory record
* and extract relevant information. There are different functions provided
* depending upon what information we need at the time. One function fills
* out an inode structure, a second one extracts a filename, a third one
* returns a symbolic link name, and a fourth one returns the extent number
* for the file.
*/
#define SIG(A,B) ((A) | ((B) << 8)) /* isonum_721() */
struct rock_state {
void *buffer;
unsigned char *chr;
int len;
int cont_size;
int cont_extent;
int cont_offset;
struct inode *inode;
};
/*
* This is a way of ensuring that we have something in the system
* use fields that is compatible with Rock Ridge. Return zero on success.
*/
static int check_sp(struct rock_ridge *rr, struct inode *inode)
{
if (rr->u.SP.magic[0] != 0xbe)
return -1;
if (rr->u.SP.magic[1] != 0xef)
return -1;
ISOFS_SB(inode->i_sb)->s_rock_offset = rr->u.SP.skip;
return 0;
}
static void setup_rock_ridge(struct iso_directory_record *de,
struct inode *inode, struct rock_state *rs)
{
rs->len = sizeof(struct iso_directory_record) + de->name_len[0];
if (rs->len & 1)
(rs->len)++;
rs->chr = (unsigned char *)de + rs->len;
rs->len = *((unsigned char *)de) - rs->len;
if (rs->len < 0)
rs->len = 0;
if (ISOFS_SB(inode->i_sb)->s_rock_offset != -1) {
rs->len -= ISOFS_SB(inode->i_sb)->s_rock_offset;
rs->chr += ISOFS_SB(inode->i_sb)->s_rock_offset;
if (rs->len < 0)
rs->len = 0;
}
}
static void init_rock_state(struct rock_state *rs, struct inode *inode)
{
memset(rs, 0, sizeof(*rs));
rs->inode = inode;
}
/*
* Returns 0 if the caller should continue scanning, 1 if the scan must end
* and -ve on error.
*/
static int rock_continue(struct rock_state *rs)
{
int ret = 1;
int blocksize = 1 << rs->inode->i_blkbits;
const int min_de_size = offsetof(struct rock_ridge, u);
kfree(rs->buffer);
rs->buffer = NULL;
if ((unsigned)rs->cont_offset > blocksize - min_de_size ||
(unsigned)rs->cont_size > blocksize ||
(unsigned)(rs->cont_offset + rs->cont_size) > blocksize) {
printk(KERN_NOTICE "rock: corrupted directory entry. "
"extent=%d, offset=%d, size=%d\n",
rs->cont_extent, rs->cont_offset, rs->cont_size);
ret = -EIO;
goto out;
}
if (rs->cont_extent) {
struct buffer_head *bh;
rs->buffer = kmalloc(rs->cont_size, GFP_KERNEL);
if (!rs->buffer) {
ret = -ENOMEM;
goto out;
}
ret = -EIO;
bh = sb_bread(rs->inode->i_sb, rs->cont_extent);
if (bh) {
memcpy(rs->buffer, bh->b_data + rs->cont_offset,
rs->cont_size);
put_bh(bh);
rs->chr = rs->buffer;
rs->len = rs->cont_size;
rs->cont_extent = 0;
rs->cont_size = 0;
rs->cont_offset = 0;
return 0;
}
printk("Unable to read rock-ridge attributes\n");
}
out:
kfree(rs->buffer);
rs->buffer = NULL;
return ret;
}
/*
* We think there's a record of type `sig' at rs->chr. Parse the signature
* and make sure that there's really room for a record of that type.
*/
static int rock_check_overflow(struct rock_state *rs, int sig)
{
int len;
switch (sig) {
case SIG('S', 'P'):
len = sizeof(struct SU_SP_s);
break;
case SIG('C', 'E'):
len = sizeof(struct SU_CE_s);
break;
case SIG('E', 'R'):
len = sizeof(struct SU_ER_s);
break;
case SIG('R', 'R'):
len = sizeof(struct RR_RR_s);
break;
case SIG('P', 'X'):
len = sizeof(struct RR_PX_s);
break;
case SIG('P', 'N'):
len = sizeof(struct RR_PN_s);
break;
case SIG('S', 'L'):
len = sizeof(struct RR_SL_s);
break;
case SIG('N', 'M'):
len = sizeof(struct RR_NM_s);
break;
case SIG('C', 'L'):
len = sizeof(struct RR_CL_s);
break;
case SIG('P', 'L'):
len = sizeof(struct RR_PL_s);
break;
case SIG('T', 'F'):
len = sizeof(struct RR_TF_s);
break;
case SIG('Z', 'F'):
len = sizeof(struct RR_ZF_s);
break;
default:
len = 0;
break;
}
len += offsetof(struct rock_ridge, u);
if (len > rs->len) {
printk(KERN_NOTICE "rock: directory entry would overflow "
"storage\n");
printk(KERN_NOTICE "rock: sig=0x%02x, size=%d, remaining=%d\n",
sig, len, rs->len);
return -EIO;
}
return 0;
}
/*
* return length of name field; 0: not found, -1: to be ignored
*/
int get_rock_ridge_filename(struct iso_directory_record *de,
char *retname, struct inode *inode)
{
struct rock_state rs;
struct rock_ridge *rr;
int sig;
int retnamlen = 0;
int truncate = 0;
int ret = 0;
if (!ISOFS_SB(inode->i_sb)->s_rock)
return 0;
*retname = 0;
init_rock_state(&rs, inode);
setup_rock_ridge(de, inode, &rs);
repeat:
while (rs.len > 2) { /* There may be one byte for padding somewhere */
rr = (struct rock_ridge *)rs.chr;
/*
* Ignore rock ridge info if rr->len is out of range, but
* don't return -EIO because that would make the file
* invisible.
*/
if (rr->len < 3)
goto out; /* Something got screwed up here */
sig = isonum_721(rs.chr);
if (rock_check_overflow(&rs, sig))
goto eio;
rs.chr += rr->len;
rs.len -= rr->len;
/*
* As above, just ignore the rock ridge info if rr->len
* is bogus.
*/
if (rs.len < 0)
goto out; /* Something got screwed up here */
switch (sig) {
case SIG('R', 'R'):
if ((rr->u.RR.flags[0] & RR_NM) == 0)
goto out;
break;
case SIG('S', 'P'):
if (check_sp(rr, inode))
goto out;
break;
case SIG('C', 'E'):
rs.cont_extent = isonum_733(rr->u.CE.extent);
rs.cont_offset = isonum_733(rr->u.CE.offset);
rs.cont_size = isonum_733(rr->u.CE.size);
break;
case SIG('N', 'M'):
if (truncate)
break;
if (rr->len < 5)
break;
/*
* If the flags are 2 or 4, this indicates '.' or '..'.
* We don't want to do anything with this, because it
* screws up the code that calls us. We don't really
* care anyways, since we can just use the non-RR
* name.
*/
if (rr->u.NM.flags & 6)
break;
if (rr->u.NM.flags & ~1) {
printk("Unsupported NM flag settings (%d)\n",
rr->u.NM.flags);
break;
}
if ((strlen(retname) + rr->len - 5) >= 254) {
truncate = 1;
break;
}
strncat(retname, rr->u.NM.name, rr->len - 5);
retnamlen += rr->len - 5;
break;
case SIG('R', 'E'):
kfree(rs.buffer);
return -1;
default:
break;
}
}
ret = rock_continue(&rs);
if (ret == 0)
goto repeat;
if (ret == 1)
return retnamlen; /* If 0, this file did not have a NM field */
out:
kfree(rs.buffer);
return ret;
eio:
ret = -EIO;
goto out;
}
static int
parse_rock_ridge_inode_internal(struct iso_directory_record *de,
struct inode *inode, int regard_xa)
{
int symlink_len = 0;
int cnt, sig;
struct inode *reloc;
struct rock_ridge *rr;
int rootflag;
struct rock_state rs;
int ret = 0;
if (!ISOFS_SB(inode->i_sb)->s_rock)
return 0;
init_rock_state(&rs, inode);
setup_rock_ridge(de, inode, &rs);
if (regard_xa) {
rs.chr += 14;
rs.len -= 14;
if (rs.len < 0)
rs.len = 0;
}
repeat:
while (rs.len > 2) { /* There may be one byte for padding somewhere */
rr = (struct rock_ridge *)rs.chr;
/*
* Ignore rock ridge info if rr->len is out of range, but
* don't return -EIO because that would make the file
* invisible.
*/
if (rr->len < 3)
goto out; /* Something got screwed up here */
sig = isonum_721(rs.chr);
if (rock_check_overflow(&rs, sig))
goto eio;
rs.chr += rr->len;
rs.len -= rr->len;
/*
* As above, just ignore the rock ridge info if rr->len
* is bogus.
*/
if (rs.len < 0)
goto out; /* Something got screwed up here */
switch (sig) {
#ifndef CONFIG_ZISOFS /* No flag for SF or ZF */
case SIG('R', 'R'):
if ((rr->u.RR.flags[0] &
(RR_PX | RR_TF | RR_SL | RR_CL)) == 0)
goto out;
break;
#endif
case SIG('S', 'P'):
if (check_sp(rr, inode))
goto out;
break;
case SIG('C', 'E'):
rs.cont_extent = isonum_733(rr->u.CE.extent);
rs.cont_offset = isonum_733(rr->u.CE.offset);
rs.cont_size = isonum_733(rr->u.CE.size);
break;
case SIG('E', 'R'):
ISOFS_SB(inode->i_sb)->s_rock = 1;
printk(KERN_DEBUG "ISO 9660 Extensions: ");
{
int p;
for (p = 0; p < rr->u.ER.len_id; p++)
printk("%c", rr->u.ER.data[p]);
}
printk("\n");
break;
case SIG('P', 'X'):
inode->i_mode = isonum_733(rr->u.PX.mode);
inode->i_nlink = isonum_733(rr->u.PX.n_links);
inode->i_uid = isonum_733(rr->u.PX.uid);
inode->i_gid = isonum_733(rr->u.PX.gid);
break;
case SIG('P', 'N'):
{
int high, low;
high = isonum_733(rr->u.PN.dev_high);
low = isonum_733(rr->u.PN.dev_low);
/*
* The Rock Ridge standard specifies that if
* sizeof(dev_t) <= 4, then the high field is
* unused, and the device number is completely
* stored in the low field. Some writers may
* ignore this subtlety,
* and as a result we test to see if the entire
* device number is
* stored in the low field, and use that.
*/
if ((low & ~0xff) && high == 0) {
inode->i_rdev =
MKDEV(low >> 8, low & 0xff);
} else {
inode->i_rdev =
MKDEV(high, low);
}
}
break;
case SIG('T', 'F'):
/*
* Some RRIP writers incorrectly place ctime in the
* TF_CREATE field. Try to handle this correctly for
* either case.
*/
/* Rock ridge never appears on a High Sierra disk */
cnt = 0;
if (rr->u.TF.flags & TF_CREATE) {
inode->i_ctime.tv_sec =
iso_date(rr->u.TF.times[cnt++].time,
0);
inode->i_ctime.tv_nsec = 0;
}
if (rr->u.TF.flags & TF_MODIFY) {
inode->i_mtime.tv_sec =
iso_date(rr->u.TF.times[cnt++].time,
0);
inode->i_mtime.tv_nsec = 0;
}
if (rr->u.TF.flags & TF_ACCESS) {
inode->i_atime.tv_sec =
iso_date(rr->u.TF.times[cnt++].time,
0);
inode->i_atime.tv_nsec = 0;
}
if (rr->u.TF.flags & TF_ATTRIBUTES) {
inode->i_ctime.tv_sec =
iso_date(rr->u.TF.times[cnt++].time,
0);
inode->i_ctime.tv_nsec = 0;
}
break;
case SIG('S', 'L'):
{
int slen;
struct SL_component *slp;
struct SL_component *oldslp;
slen = rr->len - 5;
slp = &rr->u.SL.link;
inode->i_size = symlink_len;
while (slen > 1) {
rootflag = 0;
switch (slp->flags & ~1) {
case 0:
inode->i_size +=
slp->len;
break;
case 2:
inode->i_size += 1;
break;
case 4:
inode->i_size += 2;
break;
case 8:
rootflag = 1;
inode->i_size += 1;
break;
default:
printk("Symlink component flag "
"not implemented\n");
}
slen -= slp->len + 2;
oldslp = slp;
slp = (struct SL_component *)
(((char *)slp) + slp->len + 2);
if (slen < 2) {
if (((rr->u.SL.
flags & 1) != 0)
&&
((oldslp->
flags & 1) == 0))
inode->i_size +=
1;
break;
}
/*
* If this component record isn't
* continued, then append a '/'.
*/
if (!rootflag
&& (oldslp->flags & 1) == 0)
inode->i_size += 1;
}
}
symlink_len = inode->i_size;
break;
case SIG('R', 'E'):
printk(KERN_WARNING "Attempt to read inode for "
"relocated directory\n");
goto out;
case SIG('C', 'L'):
ISOFS_I(inode)->i_first_extent =
isonum_733(rr->u.CL.location);
reloc =
isofs_iget(inode->i_sb,
ISOFS_I(inode)->i_first_extent,
0);
if (IS_ERR(reloc)) {
ret = PTR_ERR(reloc);
goto out;
}
inode->i_mode = reloc->i_mode;
inode->i_nlink = reloc->i_nlink;
inode->i_uid = reloc->i_uid;
inode->i_gid = reloc->i_gid;
inode->i_rdev = reloc->i_rdev;
inode->i_size = reloc->i_size;
inode->i_blocks = reloc->i_blocks;
inode->i_atime = reloc->i_atime;
inode->i_ctime = reloc->i_ctime;
inode->i_mtime = reloc->i_mtime;
iput(reloc);
break;
#ifdef CONFIG_ZISOFS
case SIG('Z', 'F'): {
int algo;
if (ISOFS_SB(inode->i_sb)->s_nocompress)
break;
algo = isonum_721(rr->u.ZF.algorithm);
if (algo == SIG('p', 'z')) {
int block_shift =
isonum_711(&rr->u.ZF.parms[1]);
if (block_shift > 17) {
printk(KERN_WARNING "isofs: "
"Can't handle ZF block "
"size of 2^%d\n",
block_shift);
} else {
/*
* Note: we don't change
* i_blocks here
*/
ISOFS_I(inode)->i_file_format =
isofs_file_compressed;
/*
* Parameters to compression
* algorithm (header size,
* block size)
*/
ISOFS_I(inode)->i_format_parm[0] =
isonum_711(&rr->u.ZF.parms[0]);
ISOFS_I(inode)->i_format_parm[1] =
isonum_711(&rr->u.ZF.parms[1]);
inode->i_size =
isonum_733(rr->u.ZF.
real_size);
}
} else {
printk(KERN_WARNING
"isofs: Unknown ZF compression "
"algorithm: %c%c\n",
rr->u.ZF.algorithm[0],
rr->u.ZF.algorithm[1]);
}
break;
}
#endif
default:
break;
}
}
ret = rock_continue(&rs);
if (ret == 0)
goto repeat;
if (ret == 1)
ret = 0;
out:
kfree(rs.buffer);
return ret;
eio:
ret = -EIO;
goto out;
}
static char *get_symlink_chunk(char *rpnt, struct rock_ridge *rr, char *plimit)
{
int slen;
int rootflag;
struct SL_component *oldslp;
struct SL_component *slp;
slen = rr->len - 5;
slp = &rr->u.SL.link;
while (slen > 1) {
rootflag = 0;
switch (slp->flags & ~1) {
case 0:
if (slp->len > plimit - rpnt)
return NULL;
memcpy(rpnt, slp->text, slp->len);
rpnt += slp->len;
break;
case 2:
if (rpnt >= plimit)
return NULL;
*rpnt++ = '.';
break;
case 4:
if (2 > plimit - rpnt)
return NULL;
*rpnt++ = '.';
*rpnt++ = '.';
break;
case 8:
if (rpnt >= plimit)
return NULL;
rootflag = 1;
*rpnt++ = '/';
break;
default:
printk("Symlink component flag not implemented (%d)\n",
slp->flags);
}
slen -= slp->len + 2;
oldslp = slp;
slp = (struct SL_component *)((char *)slp + slp->len + 2);
if (slen < 2) {
/*
* If there is another SL record, and this component
* record isn't continued, then add a slash.
*/
if ((!rootflag) && (rr->u.SL.flags & 1) &&
!(oldslp->flags & 1)) {
if (rpnt >= plimit)
return NULL;
*rpnt++ = '/';
}
break;
}
/*
* If this component record isn't continued, then append a '/'.
*/
if (!rootflag && !(oldslp->flags & 1)) {
if (rpnt >= plimit)
return NULL;
*rpnt++ = '/';
}
}
return rpnt;
}
int parse_rock_ridge_inode(struct iso_directory_record *de, struct inode *inode)
{
int result = parse_rock_ridge_inode_internal(de, inode, 0);
/*
* if rockridge flag was reset and we didn't look for attributes
* behind eventual XA attributes, have a look there
*/
if ((ISOFS_SB(inode->i_sb)->s_rock_offset == -1)
&& (ISOFS_SB(inode->i_sb)->s_rock == 2)) {
result = parse_rock_ridge_inode_internal(de, inode, 14);
}
return result;
}
/*
* readpage() for symlinks: reads symlink contents into the page and either
* makes it uptodate and returns 0 or returns error (-EIO)
*/
static int rock_ridge_symlink_readpage(struct file *file, struct page *page)
{
struct inode *inode = page->mapping->host;
struct iso_inode_info *ei = ISOFS_I(inode);
struct isofs_sb_info *sbi = ISOFS_SB(inode->i_sb);
char *link = kmap(page);
unsigned long bufsize = ISOFS_BUFFER_SIZE(inode);
struct buffer_head *bh;
char *rpnt = link;
unsigned char *pnt;
struct iso_directory_record *raw_de;
unsigned long block, offset;
int sig;
struct rock_ridge *rr;
struct rock_state rs;
int ret;
if (!sbi->s_rock)
goto error;
init_rock_state(&rs, inode);
block = ei->i_iget5_block;
mutex_lock(&sbi->s_mutex);
bh = sb_bread(inode->i_sb, block);
if (!bh)
goto out_noread;
offset = ei->i_iget5_offset;
pnt = (unsigned char *)bh->b_data + offset;
raw_de = (struct iso_directory_record *)pnt;
/*
* If we go past the end of the buffer, there is some sort of error.
*/
if (offset + *pnt > bufsize)
goto out_bad_span;
/*
* Now test for possible Rock Ridge extensions which will override
* some of these numbers in the inode structure.
*/
setup_rock_ridge(raw_de, inode, &rs);
repeat:
while (rs.len > 2) { /* There may be one byte for padding somewhere */
rr = (struct rock_ridge *)rs.chr;
if (rr->len < 3)
goto out; /* Something got screwed up here */
sig = isonum_721(rs.chr);
if (rock_check_overflow(&rs, sig))
goto out;
rs.chr += rr->len;
rs.len -= rr->len;
if (rs.len < 0)
goto out; /* corrupted isofs */
switch (sig) {
case SIG('R', 'R'):
if ((rr->u.RR.flags[0] & RR_SL) == 0)
goto out;
break;
case SIG('S', 'P'):
if (check_sp(rr, inode))
goto out;
break;
case SIG('S', 'L'):
rpnt = get_symlink_chunk(rpnt, rr,
link + (PAGE_SIZE - 1));
if (rpnt == NULL)
goto out;
break;
case SIG('C', 'E'):
/* This tells is if there is a continuation record */
rs.cont_extent = isonum_733(rr->u.CE.extent);
rs.cont_offset = isonum_733(rr->u.CE.offset);
rs.cont_size = isonum_733(rr->u.CE.size);
default:
break;
}
}
ret = rock_continue(&rs);
if (ret == 0)
goto repeat;
if (ret < 0)
goto fail;
if (rpnt == link)
goto fail;
brelse(bh);
*rpnt = '\0';
mutex_unlock(&sbi->s_mutex);
SetPageUptodate(page);
kunmap(page);
unlock_page(page);
return 0;
/* error exit from macro */
out:
kfree(rs.buffer);
goto fail;
out_noread:
printk("unable to read i-node block");
goto fail;
out_bad_span:
printk("symlink spans iso9660 blocks\n");
fail:
brelse(bh);
mutex_unlock(&sbi->s_mutex);
error:
SetPageError(page);
kunmap(page);
unlock_page(page);
return -EIO;
}
const struct address_space_operations isofs_symlink_aops = {
.readpage = rock_ridge_symlink_readpage
};
Reference in a new issue View git blame Copy permalink