Public/Ext4Fsd
1
0
Fork 0
mirror of https://github.com/bobranten/Ext4Fsd.git synced 2025-10-30 05:18:31 -05:00
Code Activity

jbd2 ported but not ready to use

Browse Source
This commit is contained in:
Bo Brantén
2020-02-15 23:25:43 +01:00
parent 89892ad2f3
commit e7380fe12f
12 changed files with 4621 additions and 123 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
Ext4Fsd/DIRS
View File

@@ -1 +1 @@
DIRS = nls ext3 ext4 jbd sys DIRS = nls ext3 ext4 jbd jbd2 sys

2
Ext4Fsd/ext4/ext4_csum.c
View File

@@ -151,7 +151,7 @@ static const __u32 crc32c_table[256] = {
* crc using table. * crc using table.
*/ */
static __u32 crc32c(__u32 crc, const __u8 *data, unsigned int length) __u32 crc32c(__u32 crc, const __u8 *data, unsigned int length)
{ {
while (length--) while (length--)
crc = crc32c_table[(crc ^ *data++) & 0xFFL] ^ (crc >> 8); crc = crc32c_table[(crc ^ *data++) & 0xFFL] ^ (crc >> 8);

46
Ext4Fsd/include/linux/jbd2.h
View File

@@ -33,6 +33,36 @@
#include <linux/module.h> #include <linux/module.h>
#endif #endif
#define KMEM_CACHE(__struct, __flags) kmem_cache_create(#__struct,\
sizeof(struct __struct), 0,\
(__flags), NULL)
#define kmalloc_array(num, size, flags) kmalloc((num) * (size), flags)
struct hlist_node {
struct hlist_node *next, **pprev;
};
/*
* A lockdep key is associated with each lock object. For static locks we use
* the lock address itself as the key. Dynamically allocated lock objects can
* have a statically or dynamically allocated key. Dynamically allocated lock
* keys must be registered before being used and must be unregistered before
* the key memory is freed.
*/
struct lockdep_subclass_key {
char __one_byte;
} __attribute__ ((__packed__));
#define MAX_LOCKDEP_SUBCLASSES 8UL
/* hash_entry is used to keep track of dynamically allocated keys. */
struct lock_class_key {
union {
struct hlist_node hash_entry;
struct lockdep_subclass_key subkeys[MAX_LOCKDEP_SUBCLASSES];
};
};
#define journal_oom_retry 1 #define journal_oom_retry 1
/* /*
@@ -63,7 +93,7 @@ void __jbd2_debug(int level, const char *file, const char *func,
#define jbd_debug(n, fmt, a...) \ #define jbd_debug(n, fmt, a...) \
__jbd2_debug((n), __FILE__, __func__, __LINE__, (fmt), ##a) __jbd2_debug((n), __FILE__, __func__, __LINE__, (fmt), ##a)
#else #else
#define jbd_debug(n, fmt, a) /**/ #define jbd_debug//(n, fmt, a) /**/
#endif #endif
extern void *jbd2_alloc(size_t size, gfp_t flags); extern void *jbd2_alloc(size_t size, gfp_t flags);
@@ -375,7 +405,7 @@ static inline void jbd_unlock_bh_journal_head(struct buffer_head *bh)
bit_spin_unlock(BH_JournalHead, &bh->b_state); bit_spin_unlock(BH_JournalHead, &bh->b_state);
} }
#define J_ASSERT(assert) BUG_ON(!(assert)) #define J_ASSERT(assert) /*BUG_ON(!(assert))*/
#define J_ASSERT_BH(bh, expr) J_ASSERT(expr) #define J_ASSERT_BH(bh, expr) J_ASSERT(expr)
#define J_ASSERT_JH(jh, expr) J_ASSERT(expr) #define J_ASSERT_JH(jh, expr) J_ASSERT(expr)
@@ -1542,8 +1572,8 @@ static inline bool jbd2_journal_has_csum_v2or3_feature(journal_t *j)
static inline int jbd2_journal_has_csum_v2or3(journal_t *journal) static inline int jbd2_journal_has_csum_v2or3(journal_t *journal)
{ {
WARN_ON_ONCE(jbd2_journal_has_csum_v2or3_feature(journal) && /*WARN_ON_ONCE(jbd2_journal_has_csum_v2or3_feature(journal) &&
journal->j_chksum_driver == NULL); journal->j_chksum_driver == NULL);*/
return journal->j_chksum_driver != NULL; return journal->j_chksum_driver != NULL;
} }
@@ -1599,10 +1629,12 @@ extern int jbd_blocks_per_page(struct inode *inode);
/* JBD uses a CRC32 checksum */ /* JBD uses a CRC32 checksum */
#define JBD_MAX_CHECKSUM_SIZE 4 #define JBD_MAX_CHECKSUM_SIZE 4
#if 0
__u32 crc32c(__u32 crc, const __u8 *data, unsigned int length);
static inline u32 jbd2_chksum(journal_t *journal, u32 crc, static inline u32 jbd2_chksum(journal_t *journal, u32 crc,
const void *address, unsigned int length) const void *address, unsigned int length)
{ {
#if 0
struct { struct {
struct shash_desc shash; struct shash_desc shash;
char ctx[JBD_MAX_CHECKSUM_SIZE]; char ctx[JBD_MAX_CHECKSUM_SIZE];
@@ -1620,8 +1652,10 @@ static inline u32 jbd2_chksum(journal_t *journal, u32 crc,
BUG_ON(err); BUG_ON(err);
return *(u32 *)desc.ctx; return *(u32 *)desc.ctx;
}
#endif #endif
return crc32c(crc, address, length);
}
/* Return most recent uncommitted transaction */ /* Return most recent uncommitted transaction */
static inline tid_t jbd2_get_latest_transaction(journal_t *journal) static inline tid_t jbd2_get_latest_transaction(journal_t *journal)

229
Ext4Fsd/include/linux/types.h
View File

@@ -1,114 +1,115 @@
#ifndef _LINUX_TYPES_H #ifndef _LINUX_TYPES_H
#define _LINUX_TYPES_H #define _LINUX_TYPES_H
#if _NT_TARGET_VERSION <= 0x500 #if _NT_TARGET_VERSION <= 0x500
#define _WIN2K_COMPAT_SLIST_USAGE #define _WIN2K_COMPAT_SLIST_USAGE
#endif #endif
#include <linux/config.h> #include <linux/config.h>
#include <ntifs.h> #include <ntifs.h>
#include <ntdddisk.h> #include <ntdddisk.h>
#include <windef.h> #include <windef.h>
#include <stdio.h> #include <stdio.h>
#include <stdlib.h> #include <stdlib.h>
#include <stdarg.h> #include <stdarg.h>
#include <wchar.h> #include <wchar.h>
typedef unsigned __int8 __u8; typedef unsigned __int8 __u8;
typedef signed __int8 __s8; typedef signed __int8 __s8;
typedef signed __int16 __s16; typedef signed __int16 __s16;
typedef unsigned __int16 __u16; typedef unsigned __int16 __u16;
typedef signed __int32 __s32; typedef signed __int32 __s32;
typedef unsigned __int32 __u32; typedef unsigned __int32 __u32;
typedef signed __int64 __s64; typedef signed __int64 __s64;
typedef unsigned __int64 __u64; typedef unsigned __int64 __u64;
#ifdef _MSC_VER #ifdef _MSC_VER
typedef __s16 s16; typedef __s16 s16;
typedef __u16 u16; typedef __u16 u16;
typedef __s32 s32; typedef __s32 s32;
typedef __u32 u32; typedef __u32 u32;
typedef __s64 s64; typedef __s64 s64;
typedef __u64 u64; typedef __u64 u64;
#endif #endif
#define __le16 u16 #define __le16 u16
#define __le32 u32 #define __le32 u32
#define __le64 u64 #define __le64 u64
#define bool BOOLEAN #define bool BOOLEAN
// //
// gcc special keyworks // gcc special keyworks
// //
#define __attribute__(x) #define __attribute__(x)
#define __bitwise #define __bitwise
#define __releases(x) #define __releases(x)
#ifdef _MSC_VER #ifdef _MSC_VER
#define inline __inline #define inline __inline
#endif #endif
#ifndef noinline #ifndef noinline
#define noinline #define noinline
#endif #endif
typedef __u32 __bitwise __be32; typedef __u64 __bitwise __be64;
typedef __u16 __bitwise __be16; typedef __u32 __bitwise __be32;
typedef __u16 __bitwise __be16;
#define uid_t u16
#define gid_t u16 #define uid_t u16
typedef int pid_t; #define gid_t u16
typedef unsigned __bitwise gfp_t; typedef int pid_t;
typedef unsigned __bitwise gfp_t;
typedef unsigned short umode_t; /* inode mode */
typedef unsigned short umode_t; /* inode mode */
/*
* The type used for indexing onto a disc or disc partition. /*
* If required, asm/types.h can override it and define * The type used for indexing onto a disc or disc partition.
* HAVE_SECTOR_T * If required, asm/types.h can override it and define
*/ * HAVE_SECTOR_T
typedef unsigned __int64 sector_t; */
typedef unsigned __int64 blkcnt_t; typedef unsigned __int64 sector_t;
typedef unsigned __int64 loff_t; typedef unsigned __int64 blkcnt_t;
typedef unsigned __int64 loff_t;
#define BITS_PER_LONG (32)
#define ORDER_PER_LONG (05) #define BITS_PER_LONG (32)
#define ORDER_PER_LONG (05)
#if defined(_WIN64)
typedef __int64 long_ptr_t; #if defined(_WIN64)
typedef unsigned __int64 ulong_ptr_t; typedef __int64 long_ptr_t;
# define CFS_BITS_PER_LONG (64) typedef unsigned __int64 ulong_ptr_t;
# define CFS_ORDER_PER_LONG (06) # define CFS_BITS_PER_LONG (64)
#else # define CFS_ORDER_PER_LONG (06)
typedef long long_ptr_t; #else
typedef unsigned long ulong_ptr_t; typedef long long_ptr_t;
# define CFS_BITS_PER_LONG (32) typedef unsigned long ulong_ptr_t;
# define CFS_ORDER_PER_LONG (05) # define CFS_BITS_PER_LONG (32)
#endif # define CFS_ORDER_PER_LONG (05)
#endif
//
// bit spin lock //
// // bit spin lock
//
#define __acquire(x)
#define __release(x) #define __acquire(x)
#define __release(x)
#define preempt_enable()
#define preempt_disable() #define preempt_enable()
#define preempt_disable()
//
// __FUNCTION__ issue //
// // __FUNCTION__ issue
//
#if _MSC_VER <= 1300
#define __FUNCTION__ ("jbd") #if _MSC_VER <= 1300
#endif #define __FUNCTION__ ("jbd")
#endif
#define BUG() do {DbgBreakPoint();} while(0)
#define BUG() do {DbgBreakPoint();} while(0)
#endif /* LINUX_TYPES_H */
#endif /* LINUX_TYPES_H */

7
Ext4Fsd/jbd2/MAKEFILE Normal file
View File

@@ -0,0 +1,7 @@
#
# DO NOT EDIT THIS FILE!!! Edit .\sources. if you want to add a new source
# file to this component. This file merely indirects to the real make file
# that is shared by all the driver components of the Windows NT DDK
#
!INCLUDE $(NTMAKEENV)\makefile.def

29
Ext4Fsd/jbd2/SOURCES Normal file
View File

@@ -0,0 +1,29 @@
#
# Sources for Ext2 file system driver for windows
# mattwu@163.com - http://ext2fsd.sf.net
#
# Name and type of the output file:
MAJORCOMP=ext2fsd
MINORCOMP=jbd2
TARGETNAME=jbd2
TARGETTYPE=LIBRARY
TARGETPATH=..\$(DDK_TARGET_OS)\$(DDKBUILDENV)\
#
# Visual Studio 6.0 BSC support
# VS6.0 doesn't support new-type BSC generated by WDK
#
!IFDEF BROWSER_INFO_SUPPORTED
USER_C_FLAGS = $(USER_C_FLAGS) /D__KERNEL__ /FR
!ELSE
USER_C_FLAGS = $(USER_C_FLAGS) /D__KERNEL__
!ENDIF
DRIVERTYPE=FS
INCLUDES=..\include;.;$(DRIVER_INC_PATH);
# The source code:
SOURCES=journal.c recovery.c revoke.c transaction.c

2767
Ext4Fsd/jbd2/journal.c Normal file
View File

File diff suppressed because it is too large Load Diff

1
Ext4Fsd/jbd2/readme.txt Normal file
View File

@@ -0,0 +1 @@
This is a port of jbd2 from the Linux 5.x source code, only the functions that are used by the Windows driver is included.

851
Ext4Fsd/jbd2/recovery.c Normal file
View File

@@ -0,0 +1,851 @@
// SPDX-License-Identifier: GPL-2.0+
/*
* linux/fs/jbd2/recovery.c
*
* Written by Stephen C. Tweedie <sct@redhat.com>, 1999
*
* Copyright 1999-2000 Red Hat Software --- All Rights Reserved
*
* Journal recovery routines for the generic filesystem journaling code;
* part of the ext2fs journaling system.
*/
#ifndef __KERNEL__
#include "jfs_user.h"
#else
#include <linux/time.h>
#include <linux/fs.h>
#include <linux/jbd2.h>
#include <linux/errno.h>
//#include <linux/crc32.h>
//#include <linux/blkdev.h>
#endif
/*
* Maintain information about the progress of the recovery job, so that
* the different passes can carry information between them.
*/
struct recovery_info
{
tid_t start_transaction;
tid_t end_transaction;
int nr_replays;
int nr_revokes;
int nr_revoke_hits;
};
enum passtype {PASS_SCAN, PASS_REVOKE, PASS_REPLAY};
static int do_one_pass(journal_t *journal,
struct recovery_info *info, enum passtype pass);
static int scan_revoke_records(journal_t *, struct buffer_head *,
tid_t, struct recovery_info *);
#ifdef __KERNEL__
/* Release readahead buffers after use */
static void journal_brelse_array(struct buffer_head *b[], int n)
{
while (--n >= 0)
brelse (b[n]);
}
/*
* When reading from the journal, we are going through the block device
* layer directly and so there is no readahead being done for us. We
* need to implement any readahead ourselves if we want it to happen at
* all. Recovery is basically one long sequential read, so make sure we
* do the IO in reasonably large chunks.
*
* This is not so critical that we need to be enormously clever about
* the readahead size, though. 128K is a purely arbitrary, good-enough
* fixed value.
*/
#define MAXBUF 8
static int do_readahead(journal_t *journal, unsigned int start)
{
int err;
unsigned int max, nbufs, next;
unsigned long long blocknr;
struct buffer_head *bh;
struct buffer_head * bufs[MAXBUF];
/* Do up to 128K of readahead */
max = start + (128 * 1024 / journal->j_blocksize);
if (max > journal->j_maxlen)
max = journal->j_maxlen;
/* Do the readahead itself. We'll submit MAXBUF buffer_heads at
* a time to the block device IO layer. */
nbufs = 0;
for (next = start; next < max; next++) {
err = jbd2_journal_bmap(journal, next, &blocknr);
if (err) {
printk(KERN_ERR "JBD2: bad block at offset %u\n",
next);
goto failed;
}
bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize);
if (!bh) {
err = -ENOMEM;
goto failed;
}
if (!buffer_uptodate(bh) && !buffer_locked(bh)) {
bufs[nbufs++] = bh;
if (nbufs == MAXBUF) {
//ll_rw_block(REQ_OP_READ, 0, nbufs, bufs);
ll_rw_block(READ, nbufs, bufs);
journal_brelse_array(bufs, nbufs);
nbufs = 0;
}
} else
brelse(bh);
}
if (nbufs)
//ll_rw_block(REQ_OP_READ, 0, nbufs, bufs);
ll_rw_block(READ, nbufs, bufs);
err = 0;
failed:
if (nbufs)
journal_brelse_array(bufs, nbufs);
return err;
}
#endif /* __KERNEL__ */
/*
* Read a block from the journal
*/
static int jread(struct buffer_head **bhp, journal_t *journal,
unsigned int offset)
{
int err;
unsigned long long blocknr;
struct buffer_head *bh;
*bhp = NULL;
if (offset >= journal->j_maxlen) {
printk(KERN_ERR "JBD2: corrupted journal superblock\n");
return -EFSCORRUPTED;
}
err = jbd2_journal_bmap(journal, offset, &blocknr);
if (err) {
printk(KERN_ERR "JBD2: bad block at offset %u\n",
offset);
return err;
}
bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize);
if (!bh)
return -ENOMEM;
if (!buffer_uptodate(bh)) {
/* If this is a brand new buffer, start readahead.
Otherwise, we assume we are already reading it. */
if (!buffer_req(bh))
do_readahead(journal, offset);
wait_on_buffer(bh);
}
if (!buffer_uptodate(bh)) {
printk(KERN_ERR "JBD2: Failed to read block at offset %u\n",
offset);
brelse(bh);
return -EIO;
}
*bhp = bh;
return 0;
}
static int jbd2_descriptor_block_csum_verify(journal_t *j, void *buf)
{
struct jbd2_journal_block_tail *tail;
__be32 provided;
__u32 calculated;
if (!jbd2_journal_has_csum_v2or3(j))
return 1;
tail = (struct jbd2_journal_block_tail *)((char*)buf + j->j_blocksize -
sizeof(struct jbd2_journal_block_tail));
provided = tail->t_checksum;
tail->t_checksum = 0;
calculated = jbd2_chksum(j, j->j_csum_seed, buf, j->j_blocksize);
tail->t_checksum = provided;
return provided == cpu_to_be32(calculated);
}
/*
* Count the number of in-use tags in a journal descriptor block.
*/
static int count_tags(journal_t *journal, struct buffer_head *bh)
{
char * tagp;
journal_block_tag_t * tag;
int nr = 0, size = journal->j_blocksize;
int tag_bytes = journal_tag_bytes(journal);
if (jbd2_journal_has_csum_v2or3(journal))
size -= sizeof(struct jbd2_journal_block_tail);
tagp = &bh->b_data[sizeof(journal_header_t)];
while ((tagp - bh->b_data + tag_bytes) <= size) {
tag = (journal_block_tag_t *) tagp;
nr++;
tagp += tag_bytes;
if (!(tag->t_flags & cpu_to_be16(JBD2_FLAG_SAME_UUID)))
tagp += 16;
if (tag->t_flags & cpu_to_be16(JBD2_FLAG_LAST_TAG))
break;
}
return nr;
}
/* Make sure we wrap around the log correctly! */
#define wrap(journal, var) \
do { \
if (var >= (journal)->j_last) \
var -= ((journal)->j_last - (journal)->j_first); \
} while (0)
/**
* jbd2_journal_recover - recovers a on-disk journal
* @journal: the journal to recover
*
* The primary function for recovering the log contents when mounting a
* journaled device.
*
* Recovery is done in three passes. In the first pass, we look for the
* end of the log. In the second, we assemble the list of revoke
* blocks. In the third and final pass, we replay any un-revoked blocks
* in the log.
*/
int jbd2_journal_recover(journal_t *journal)
{
int err, err2;
journal_superblock_t * sb;
struct recovery_info info;
memset(&info, 0, sizeof(info));
sb = journal->j_superblock;
/*
* The journal superblock's s_start field (the current log head)
* is always zero if, and only if, the journal was cleanly
* unmounted.
*/
if (!sb->s_start) {
jbd_debug(1, "No recovery required, last transaction %d\n",
be32_to_cpu(sb->s_sequence));
journal->j_transaction_sequence = be32_to_cpu(sb->s_sequence) + 1;
return 0;
}
err = do_one_pass(journal, &info, PASS_SCAN);
if (!err)
err = do_one_pass(journal, &info, PASS_REVOKE);
if (!err)
err = do_one_pass(journal, &info, PASS_REPLAY);
jbd_debug(1, "JBD2: recovery, exit status %d, "
"recovered transactions %u to %u\n",
err, info.start_transaction, info.end_transaction);
jbd_debug(1, "JBD2: Replayed %d and revoked %d/%d blocks\n",
info.nr_replays, info.nr_revoke_hits, info.nr_revokes);
/* Restart the log at the next transaction ID, thus invalidating
* any existing commit records in the log. */
journal->j_transaction_sequence = ++info.end_transaction;
jbd2_journal_clear_revoke(journal);
err2 = sync_blockdev(journal->j_fs_dev);
if (!err)
err = err2;
/* Make sure all replayed data is on permanent storage */
/*if (journal->j_flags & JBD2_BARRIER) {
err2 = blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL);
if (!err)
err = err2;
}*/
return err;
}
#if 0
/**
* jbd2_journal_skip_recovery - Start journal and wipe exiting records
* @journal: journal to startup
*
* Locate any valid recovery information from the journal and set up the
* journal structures in memory to ignore it (presumably because the
* caller has evidence that it is out of date).
* This function doesn't appear to be exported..
*
* We perform one pass over the journal to allow us to tell the user how
* much recovery information is being erased, and to let us initialise
* the journal transaction sequence numbers to the next unused ID.
*/
int jbd2_journal_skip_recovery(journal_t *journal)
{
int err;
struct recovery_info info;
memset (&info, 0, sizeof(info));
err = do_one_pass(journal, &info, PASS_SCAN);
if (err) {
printk(KERN_ERR "JBD2: error %d scanning journal\n", err);
++journal->j_transaction_sequence;
} else {
#ifdef CONFIG_JBD2_DEBUG
int dropped = info.end_transaction -
be32_to_cpu(journal->j_superblock->s_sequence);
jbd_debug(1,
"JBD2: ignoring %d transaction%s from the journal.\n",
dropped, (dropped == 1) ? "" : "s");
#endif
journal->j_transaction_sequence = ++info.end_transaction;
}
journal->j_tail = 0;
return err;
}
#endif
static inline unsigned long long read_tag_block(journal_t *journal,
journal_block_tag_t *tag)
{
unsigned long long block = be32_to_cpu(tag->t_blocknr);
if (jbd2_has_feature_64bit(journal))
block |= (u64)be32_to_cpu(tag->t_blocknr_high) << 32;
return block;
}
#define crc32_be crc32c
/*
* calc_chksums calculates the checksums for the blocks described in the
* descriptor block.
*/
static int calc_chksums(journal_t *journal, struct buffer_head *bh,
unsigned long *next_log_block, __u32 *crc32_sum)
{
int i, num_blks, err;
unsigned long io_block;
struct buffer_head *obh;
num_blks = count_tags(journal, bh);
/* Calculate checksum of the descriptor block. */
*crc32_sum = crc32_be(*crc32_sum, (void *)bh->b_data, bh->b_size);
for (i = 0; i < num_blks; i++) {
io_block = (*next_log_block)++;
wrap(journal, *next_log_block);
err = jread(&obh, journal, io_block);
if (err) {
printk(KERN_ERR "JBD2: IO error %d recovering block "
"%lu in log\n", err, io_block);
return 1;
} else {
*crc32_sum = crc32_be(*crc32_sum, (void *)obh->b_data,
obh->b_size);
}
put_bh(obh);
}
return 0;
}
static int jbd2_commit_block_csum_verify(journal_t *j, void *buf)
{
struct commit_header *h;
__be32 provided;
__u32 calculated;
if (!jbd2_journal_has_csum_v2or3(j))
return 1;
h = buf;
provided = h->h_chksum[0];
h->h_chksum[0] = 0;
calculated = jbd2_chksum(j, j->j_csum_seed, buf, j->j_blocksize);
h->h_chksum[0] = provided;
return provided == cpu_to_be32(calculated);
}
static int jbd2_block_tag_csum_verify(journal_t *j, journal_block_tag_t *tag,
void *buf, __u32 sequence)
{
journal_block_tag3_t *tag3 = (journal_block_tag3_t *)tag;
__u32 csum32;
__be32 seq;
if (!jbd2_journal_has_csum_v2or3(j))
return 1;
seq = cpu_to_be32(sequence);
csum32 = jbd2_chksum(j, j->j_csum_seed, (__u8 *)&seq, sizeof(seq));
csum32 = jbd2_chksum(j, csum32, buf, j->j_blocksize);
if (jbd2_has_feature_csum3(j))
return tag3->t_checksum == cpu_to_be32(csum32);
else
return tag->t_checksum == cpu_to_be16(csum32);
}
static int do_one_pass(journal_t *journal,
struct recovery_info *info, enum passtype pass)
{
unsigned int first_commit_ID, next_commit_ID;
unsigned long next_log_block;
int err, success = 0;
journal_superblock_t * sb;
journal_header_t * tmp;
struct buffer_head * bh;
unsigned int sequence;
int blocktype;
int tag_bytes = journal_tag_bytes(journal);
__u32 crc32_sum = ~0; /* Transactional Checksums */
int descr_csum_size = 0;
int block_error = 0;
/*
* First thing is to establish what we expect to find in the log
* (in terms of transaction IDs), and where (in terms of log
* block offsets): query the superblock.
*/
sb = journal->j_superblock;
next_commit_ID = be32_to_cpu(sb->s_sequence);
next_log_block = be32_to_cpu(sb->s_start);
first_commit_ID = next_commit_ID;
if (pass == PASS_SCAN)
info->start_transaction = first_commit_ID;
jbd_debug(1, "Starting recovery pass %d\n", pass);
/*
* Now we walk through the log, transaction by transaction,
* making sure that each transaction has a commit block in the
* expected place. Each complete transaction gets replayed back
* into the main filesystem.
*/
while (1) {
int flags;
char * tagp;
journal_block_tag_t * tag;
struct buffer_head * obh;
struct buffer_head * nbh;
cond_resched();
/* If we already know where to stop the log traversal,
* check right now that we haven't gone past the end of
* the log. */
if (pass != PASS_SCAN)
if (tid_geq(next_commit_ID, info->end_transaction))
break;
jbd_debug(2, "Scanning for sequence ID %u at %lu/%lu\n",
next_commit_ID, next_log_block, journal->j_last);
/* Skip over each chunk of the transaction looking
* either the next descriptor block or the final commit
* record. */
jbd_debug(3, "JBD2: checking block %ld\n", next_log_block);
err = jread(&bh, journal, next_log_block);
if (err)
goto failed;
next_log_block++;
wrap(journal, next_log_block);
/* What kind of buffer is it?
*
* If it is a descriptor block, check that it has the
* expected sequence number. Otherwise, we're all done
* here. */
tmp = (journal_header_t *)bh->b_data;
if (tmp->h_magic != cpu_to_be32(JBD2_MAGIC_NUMBER)) {
brelse(bh);
break;
}
blocktype = be32_to_cpu(tmp->h_blocktype);
sequence = be32_to_cpu(tmp->h_sequence);
jbd_debug(3, "Found magic %d, sequence %d\n",
blocktype, sequence);
if (sequence != next_commit_ID) {
brelse(bh);
break;
}
/* OK, we have a valid descriptor block which matches
* all of the sequence number checks. What are we going
* to do with it? That depends on the pass... */
switch(blocktype) {
case JBD2_DESCRIPTOR_BLOCK:
/* Verify checksum first */
if (jbd2_journal_has_csum_v2or3(journal))
descr_csum_size =
sizeof(struct jbd2_journal_block_tail);
if (descr_csum_size > 0 &&
!jbd2_descriptor_block_csum_verify(journal,
bh->b_data)) {
printk(KERN_ERR "JBD2: Invalid checksum "
"recovering block %lu in log\n",
next_log_block);
err = -EFSBADCRC;
brelse(bh);
goto failed;
}
/* If it is a valid descriptor block, replay it
* in pass REPLAY; if journal_checksums enabled, then
* calculate checksums in PASS_SCAN, otherwise,
* just skip over the blocks it describes. */
if (pass != PASS_REPLAY) {
if (pass == PASS_SCAN &&
jbd2_has_feature_checksum(journal) &&
!info->end_transaction) {
if (calc_chksums(journal, bh,
&next_log_block,
&crc32_sum)) {
put_bh(bh);
break;
}
put_bh(bh);
continue;
}
next_log_block += count_tags(journal, bh);
wrap(journal, next_log_block);
put_bh(bh);
continue;
}
/* A descriptor block: we can now write all of
* the data blocks. Yay, useful work is finally
* getting done here! */
tagp = &bh->b_data[sizeof(journal_header_t)];
while ((tagp - bh->b_data + tag_bytes)
<= journal->j_blocksize - descr_csum_size) {
unsigned long io_block;
tag = (journal_block_tag_t *) tagp;
flags = be16_to_cpu(tag->t_flags);
io_block = next_log_block++;
wrap(journal, next_log_block);
err = jread(&obh, journal, io_block);
if (err) {
/* Recover what we can, but
* report failure at the end. */
success = err;
printk(KERN_ERR
"JBD2: IO error %d recovering "
"block %ld in log\n",
err, io_block);
} else {
unsigned long long blocknr;
J_ASSERT(obh != NULL);
blocknr = read_tag_block(journal,
tag);
/* If the block has been
* revoked, then we're all done
* here. */
if (jbd2_journal_test_revoke
(journal, blocknr,
next_commit_ID)) {
brelse(obh);
++info->nr_revoke_hits;
goto skip_write;
}
/* Look for block corruption */
if (!jbd2_block_tag_csum_verify(
journal, tag, obh->b_data,
be32_to_cpu(tmp->h_sequence))) {
brelse(obh);
success = -EFSBADCRC;
printk(KERN_ERR "JBD2: Invalid "
"checksum recovering "
"data block %llu in "
"log\n", blocknr);
block_error = 1;
goto skip_write;
}
/* Find a buffer for the new
* data being restored */
nbh = __getblk(journal->j_fs_dev,
blocknr,
journal->j_blocksize);
if (nbh == NULL) {
printk(KERN_ERR
"JBD2: Out of memory "
"during recovery.\n");
err = -ENOMEM;
brelse(bh);
brelse(obh);
goto failed;
}
lock_buffer(nbh);
memcpy(nbh->b_data, obh->b_data,
journal->j_blocksize);
if (flags & JBD2_FLAG_ESCAPE) {
*((__be32 *)nbh->b_data) =
cpu_to_be32(JBD2_MAGIC_NUMBER);
}
BUFFER_TRACE(nbh, "marking dirty");
set_buffer_uptodate(nbh);
mark_buffer_dirty(nbh);
BUFFER_TRACE(nbh, "marking uptodate");
++info->nr_replays;
/* ll_rw_block(WRITE, 1, &nbh); */
unlock_buffer(nbh);
brelse(obh);
brelse(nbh);
}
skip_write:
tagp += tag_bytes;
if (!(flags & JBD2_FLAG_SAME_UUID))
tagp += 16;
if (flags & JBD2_FLAG_LAST_TAG)
break;
}
brelse(bh);
continue;
case JBD2_COMMIT_BLOCK:
/* How to differentiate between interrupted commit
* and journal corruption ?
*
* {nth transaction}
* Checksum Verification Failed
* |
* ____________________
* | |
* async_commit sync_commit
* | |
* | GO TO NEXT "Journal Corruption"
* | TRANSACTION
* |
* {(n+1)th transanction}
* |
* _______|______________
* | |
* Commit block found Commit block not found
* | |
* "Journal Corruption" |
* _____________|_________
* | |
* nth trans corrupt OR nth trans
* and (n+1)th interrupted interrupted
* before commit block
* could reach the disk.
* (Cannot find the difference in above
* mentioned conditions. Hence assume
* "Interrupted Commit".)
*/
/* Found an expected commit block: if checksums
* are present verify them in PASS_SCAN; else not
* much to do other than move on to the next sequence
* number. */
if (pass == PASS_SCAN &&
jbd2_has_feature_checksum(journal)) {
int chksum_err, chksum_seen;
struct commit_header *cbh =
(struct commit_header *)bh->b_data;
unsigned found_chksum =
be32_to_cpu(cbh->h_chksum[0]);
chksum_err = chksum_seen = 0;
if (info->end_transaction) {
journal->j_failed_commit =
info->end_transaction;
brelse(bh);
break;
}
if (crc32_sum == found_chksum &&
cbh->h_chksum_type == JBD2_CRC32_CHKSUM &&
cbh->h_chksum_size ==
JBD2_CRC32_CHKSUM_SIZE)
chksum_seen = 1;
else if (!(cbh->h_chksum_type == 0 &&
cbh->h_chksum_size == 0 &&
found_chksum == 0 &&
!chksum_seen))
/*
* If fs is mounted using an old kernel and then
* kernel with journal_chksum is used then we
* get a situation where the journal flag has
* checksum flag set but checksums are not
* present i.e chksum = 0, in the individual
* commit blocks.
* Hence to avoid checksum failures, in this
* situation, this extra check is added.
*/
chksum_err = 1;
if (chksum_err) {
info->end_transaction = next_commit_ID;
if (!jbd2_has_feature_async_commit(journal)) {
journal->j_failed_commit =
next_commit_ID;
brelse(bh);
break;
}
}
crc32_sum = ~0;
}
if (pass == PASS_SCAN &&
!jbd2_commit_block_csum_verify(journal,
bh->b_data)) {
info->end_transaction = next_commit_ID;
if (!jbd2_has_feature_async_commit(journal)) {
journal->j_failed_commit =
next_commit_ID;
brelse(bh);
break;
}
}
brelse(bh);
next_commit_ID++;
continue;
case JBD2_REVOKE_BLOCK:
/* If we aren't in the REVOKE pass, then we can
* just skip over this block. */
if (pass != PASS_REVOKE) {
brelse(bh);
continue;
}
err = scan_revoke_records(journal, bh,
next_commit_ID, info);
brelse(bh);
if (err)
goto failed;
continue;
default:
jbd_debug(3, "Unrecognised magic %d, end of scan.\n",
blocktype);
brelse(bh);
goto done;
}
}
done:
/*
* We broke out of the log scan loop: either we came to the
* known end of the log or we found an unexpected block in the
* log. If the latter happened, then we know that the "current"
* transaction marks the end of the valid log.
*/
if (pass == PASS_SCAN) {
if (!info->end_transaction)
info->end_transaction = next_commit_ID;
} else {
/* It's really bad news if different passes end up at
* different places (but possible due to IO errors). */
if (info->end_transaction != next_commit_ID) {
printk(KERN_ERR "JBD2: recovery pass %d ended at "
"transaction %u, expected %u\n",
pass, next_commit_ID, info->end_transaction);
if (!success)
success = -EIO;
}
}
if (block_error && success == 0)
success = -EIO;
return success;
failed:
return err;
}
/* Scan a revoke record, marking all blocks mentioned as revoked. */
static int scan_revoke_records(journal_t *journal, struct buffer_head *bh,
tid_t sequence, struct recovery_info *info)
{
jbd2_journal_revoke_header_t *header;
int offset, max;
unsigned int csum_size = 0;
__u32 rcount;
int record_len = 4;
header = (jbd2_journal_revoke_header_t *) bh->b_data;
offset = sizeof(jbd2_journal_revoke_header_t);
rcount = be32_to_cpu(header->r_count);
if (!jbd2_descriptor_block_csum_verify(journal, header))
return -EFSBADCRC;
if (jbd2_journal_has_csum_v2or3(journal))
csum_size = sizeof(struct jbd2_journal_block_tail);
if (rcount > journal->j_blocksize - csum_size)
return -EINVAL;
max = rcount;
if (jbd2_has_feature_64bit(journal))
record_len = 8;
while (offset + record_len <= max) {
unsigned long long blocknr;
int err;
if (record_len == 4)
blocknr = be32_to_cpu(* ((__be32 *) (bh->b_data+offset)));
else
blocknr = be64_to_cpu(* ((__be64 *) (bh->b_data+offset)));
offset += record_len;
err = jbd2_journal_set_revoke(journal, blocknr, sequence);
if (err)
return err;
++info->nr_revokes;
}
return 0;
}

755
Ext4Fsd/jbd2/revoke.c Normal file
View File

@@ -0,0 +1,755 @@
// SPDX-License-Identifier: GPL-2.0+
/*
* linux/fs/jbd2/revoke.c
*
* Written by Stephen C. Tweedie <sct@redhat.com>, 2000
*
* Copyright 2000 Red Hat corp --- All Rights Reserved
*
* Journal revoke routines for the generic filesystem journaling code;
* part of the ext2fs journaling system.
*
* Revoke is the mechanism used to prevent old log records for deleted
* metadata from being replayed on top of newer data using the same
* blocks. The revoke mechanism is used in two separate places:
*
* + Commit: during commit we write the entire list of the current
* transaction's revoked blocks to the journal
*
* + Recovery: during recovery we record the transaction ID of all
* revoked blocks. If there are multiple revoke records in the log
* for a single block, only the last one counts, and if there is a log
* entry for a block beyond the last revoke, then that log entry still
* gets replayed.
*
* We can get interactions between revokes and new log data within a
* single transaction:
*
* Block is revoked and then journaled:
* The desired end result is the journaling of the new block, so we
* cancel the revoke before the transaction commits.
*
* Block is journaled and then revoked:
* The revoke must take precedence over the write of the block, so we
* need either to cancel the journal entry or to write the revoke
* later in the log than the log block. In this case, we choose the
* latter: journaling a block cancels any revoke record for that block
* in the current transaction, so any revoke for that block in the
* transaction must have happened after the block was journaled and so
* the revoke must take precedence.
*
* Block is revoked and then written as data:
* The data write is allowed to succeed, but the revoke is _not_
* cancelled. We still need to prevent old log records from
* overwriting the new data. We don't even need to clear the revoke
* bit here.
*
* We cache revoke status of a buffer in the current transaction in b_states
* bits. As the name says, revokevalid flag indicates that the cached revoke
* status of a buffer is valid and we can rely on the cached status.
*
* Revoke information on buffers is a tri-state value:
*
* RevokeValid clear: no cached revoke status, need to look it up
* RevokeValid set, Revoked clear:
* buffer has not been revoked, and cancel_revoke
* need do nothing.
* RevokeValid set, Revoked set:
* buffer has been revoked.
*
* Locking rules:
* We keep two hash tables of revoke records. One hashtable belongs to the
* running transaction (is pointed to by journal->j_revoke), the other one
* belongs to the committing transaction. Accesses to the second hash table
* happen only from the kjournald and no other thread touches this table. Also
* journal_switch_revoke_table() which switches which hashtable belongs to the
* running and which to the committing transaction is called only from
* kjournald. Therefore we need no locks when accessing the hashtable belonging
* to the committing transaction.
*
* All users operating on the hash table belonging to the running transaction
* have a handle to the transaction. Therefore they are safe from kjournald
* switching hash tables under them. For operations on the lists of entries in
* the hash table j_revoke_lock is used.
*
* Finally, also replay code uses the hash tables but at this moment no one else
* can touch them (filesystem isn't mounted yet) and hence no locking is
* needed.
*/
#ifndef __KERNEL__
#include "jfs_user.h"
#else
#include <linux/time.h>
#include <linux/fs.h>
#include <linux/jbd2.h>
#include <linux/errno.h>
#include <linux/slab.h>
#include <linux/list.h>
#include <linux/init.h>
//#include <linux/bio.h>
#include <linux/log2.h>
//#include <linux/hash.h>
#endif
static struct kmem_cache *jbd2_revoke_record_cache;
static struct kmem_cache *jbd2_revoke_table_cache;
/* Each revoke record represents one single revoked block. During
journal replay, this involves recording the transaction ID of the
last transaction to revoke this block. */
struct jbd2_revoke_record_s
{
struct list_head hash;
tid_t sequence; /* Used for recovery only */
unsigned long long blocknr;
};
/* The revoke table is just a simple hash table of revoke records. */
struct jbd2_revoke_table_s
{
/* It is conceivable that we might want a larger hash table
* for recovery. Must be a power of two. */
int hash_size;
int hash_shift;
struct list_head *hash_table;
};
#if 0
#ifdef __KERNEL__
static void write_one_revoke_record(transaction_t *,
struct list_head *,
struct buffer_head **, int *,
struct jbd2_revoke_record_s *);
static void flush_descriptor(journal_t *, struct buffer_head *, int);
#endif
#endif
/* Utility functions to maintain the revoke table */
#define GOLDEN_RATIO_32 0x61C88647
#define GOLDEN_RATIO_64 0x61C8864680B583EBull
static inline u32 __hash_32(u32 val)
{
return val * GOLDEN_RATIO_32;
}
static inline u32 hash_32(u32 val, unsigned int bits)
{
/* High bits are more random, so use them. */
return __hash_32(val) >> (32 - bits);
}
static inline u32 hash_64(u64 val, unsigned int bits)
{
#if BITS_PER_LONG == 64
/* 64x64-bit multiply is efficient on all 64-bit processors */
return val * GOLDEN_RATIO_64 >> (64 - bits);
#else
/* Hash 64 bits using only 32x32-bit multiply. */
return hash_32((u32)val ^ __hash_32(val >> 32), bits);
#endif
}
static inline int hash(journal_t *journal, unsigned long long block)
{
return hash_64(block, journal->j_revoke->hash_shift);
}
static int insert_revoke_hash(journal_t *journal, unsigned long long blocknr,
tid_t seq)
{
struct list_head *hash_list;
struct jbd2_revoke_record_s *record;
gfp_t gfp_mask = GFP_NOFS;
if (journal_oom_retry)
gfp_mask |= __GFP_NOFAIL;
record = kmem_cache_alloc(jbd2_revoke_record_cache, gfp_mask);
if (!record)
return -ENOMEM;
record->sequence = seq;
record->blocknr = blocknr;
hash_list = &journal->j_revoke->hash_table[hash(journal, blocknr)];
spin_lock(&journal->j_revoke_lock);
list_add(&record->hash, hash_list);
spin_unlock(&journal->j_revoke_lock);
return 0;
}
/* Find a revoke record in the journal's hash table. */
static struct jbd2_revoke_record_s *find_revoke_record(journal_t *journal,
unsigned long long blocknr)
{
struct list_head *hash_list;
struct jbd2_revoke_record_s *record;
hash_list = &journal->j_revoke->hash_table[hash(journal, blocknr)];
spin_lock(&journal->j_revoke_lock);
record = (struct jbd2_revoke_record_s *) hash_list->next;
while (&(record->hash) != hash_list) {
if (record->blocknr == blocknr) {
spin_unlock(&journal->j_revoke_lock);
return record;
}
record = (struct jbd2_revoke_record_s *) record->hash.next;
}
spin_unlock(&journal->j_revoke_lock);
return NULL;
}
void jbd2_journal_destroy_revoke_caches(void)
{
kmem_cache_destroy(jbd2_revoke_record_cache);
jbd2_revoke_record_cache = NULL;
kmem_cache_destroy(jbd2_revoke_table_cache);
jbd2_revoke_table_cache = NULL;
}
int __init jbd2_journal_init_revoke_caches(void)
{
J_ASSERT(!jbd2_revoke_record_cache);
J_ASSERT(!jbd2_revoke_table_cache);
jbd2_revoke_record_cache = KMEM_CACHE(jbd2_revoke_record_s,
SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY);
if (!jbd2_revoke_record_cache)
goto record_cache_failure;
jbd2_revoke_table_cache = KMEM_CACHE(jbd2_revoke_table_s,
SLAB_TEMPORARY);
if (!jbd2_revoke_table_cache)
goto table_cache_failure;
return 0;
table_cache_failure:
jbd2_journal_destroy_revoke_caches();
record_cache_failure:
return -ENOMEM;
}
static struct jbd2_revoke_table_s *jbd2_journal_init_revoke_table(int hash_size)
{
int shift = 0;
int tmp = hash_size;
struct jbd2_revoke_table_s *table;
table = kmem_cache_alloc(jbd2_revoke_table_cache, GFP_KERNEL);
if (!table)
goto out;
while((tmp >>= 1UL) != 0UL)
shift++;
table->hash_size = hash_size;
table->hash_shift = shift;
table->hash_table =
kmalloc_array(hash_size, sizeof(struct list_head), GFP_KERNEL);
if (!table->hash_table) {
kmem_cache_free(jbd2_revoke_table_cache, table);
table = NULL;
goto out;
}
for (tmp = 0; tmp < hash_size; tmp++)
INIT_LIST_HEAD(&table->hash_table[tmp]);
out:
return table;
}
static void jbd2_journal_destroy_revoke_table(struct jbd2_revoke_table_s *table)
{
int i;
struct list_head *hash_list;
for (i = 0; i < table->hash_size; i++) {
hash_list = &table->hash_table[i];
J_ASSERT(list_empty(hash_list));
}
kfree(table->hash_table);
kmem_cache_free(jbd2_revoke_table_cache, table);
}
/* Initialise the revoke table for a given journal to a given size. */
int jbd2_journal_init_revoke(journal_t *journal, int hash_size)
{
J_ASSERT(journal->j_revoke_table[0] == NULL);
J_ASSERT(is_power_of_2(hash_size));
journal->j_revoke_table[0] = jbd2_journal_init_revoke_table(hash_size);
if (!journal->j_revoke_table[0])
goto fail0;
journal->j_revoke_table[1] = jbd2_journal_init_revoke_table(hash_size);
if (!journal->j_revoke_table[1])
goto fail1;
journal->j_revoke = journal->j_revoke_table[1];
spin_lock_init(&journal->j_revoke_lock);
return 0;
fail1:
jbd2_journal_destroy_revoke_table(journal->j_revoke_table[0]);
journal->j_revoke_table[0] = NULL;
fail0:
return -ENOMEM;
}
/* Destroy a journal's revoke table. The table must already be empty! */
void jbd2_journal_destroy_revoke(journal_t *journal)
{
journal->j_revoke = NULL;
if (journal->j_revoke_table[0])
jbd2_journal_destroy_revoke_table(journal->j_revoke_table[0]);
if (journal->j_revoke_table[1])
jbd2_journal_destroy_revoke_table(journal->j_revoke_table[1]);
}
#if 0
#ifdef __KERNEL__
/*
* jbd2_journal_revoke: revoke a given buffer_head from the journal. This
* prevents the block from being replayed during recovery if we take a
* crash after this current transaction commits. Any subsequent
* metadata writes of the buffer in this transaction cancel the
* revoke.
*
* Note that this call may block --- it is up to the caller to make
* sure that there are no further calls to journal_write_metadata
* before the revoke is complete. In ext3, this implies calling the
* revoke before clearing the block bitmap when we are deleting
* metadata.
*
* Revoke performs a jbd2_journal_forget on any buffer_head passed in as a
* parameter, but does _not_ forget the buffer_head if the bh was only
* found implicitly.
*
* bh_in may not be a journalled buffer - it may have come off
* the hash tables without an attached journal_head.
*
* If bh_in is non-zero, jbd2_journal_revoke() will decrement its b_count
* by one.
*/
int jbd2_journal_revoke(handle_t *handle, unsigned long long blocknr,
struct buffer_head *bh_in)
{
struct buffer_head *bh = NULL;
journal_t *journal;
struct block_device *bdev;
int err;
might_sleep();
if (bh_in)
BUFFER_TRACE(bh_in, "enter");
journal = handle->h_transaction->t_journal;
if (!jbd2_journal_set_features(journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)){
J_ASSERT (!"Cannot set revoke feature!");
return -EINVAL;
}
bdev = journal->j_fs_dev;
bh = bh_in;
if (!bh) {
bh = __find_get_block(bdev, blocknr, journal->j_blocksize);
if (bh)
BUFFER_TRACE(bh, "found on hash");
}
#ifdef JBD2_EXPENSIVE_CHECKING
else {
struct buffer_head *bh2;
/* If there is a different buffer_head lying around in
* memory anywhere... */
bh2 = __find_get_block(bdev, blocknr, journal->j_blocksize);
if (bh2) {
/* ... and it has RevokeValid status... */
if (bh2 != bh && buffer_revokevalid(bh2))
/* ...then it better be revoked too,
* since it's illegal to create a revoke
* record against a buffer_head which is
* not marked revoked --- that would
* risk missing a subsequent revoke
* cancel. */
J_ASSERT_BH(bh2, buffer_revoked(bh2));
put_bh(bh2);
}
}
#endif
/* We really ought not ever to revoke twice in a row without
first having the revoke cancelled: it's illegal to free a
block twice without allocating it in between! */
if (bh) {
if (!J_EXPECT_BH(bh, !buffer_revoked(bh),
"inconsistent data on disk")) {
if (!bh_in)
brelse(bh);
return -EIO;
}
set_buffer_revoked(bh);
set_buffer_revokevalid(bh);
if (bh_in) {
BUFFER_TRACE(bh_in, "call jbd2_journal_forget");
jbd2_journal_forget(handle, bh_in);
} else {
BUFFER_TRACE(bh, "call brelse");
__brelse(bh);
}
}
jbd_debug(2, "insert revoke for block %llu, bh_in=%p\n",blocknr, bh_in);
err = insert_revoke_hash(journal, blocknr,
handle->h_transaction->t_tid);
BUFFER_TRACE(bh_in, "exit");
return err;
}
/*
* Cancel an outstanding revoke. For use only internally by the
* journaling code (called from jbd2_journal_get_write_access).
*
* We trust buffer_revoked() on the buffer if the buffer is already
* being journaled: if there is no revoke pending on the buffer, then we
* don't do anything here.
*
* This would break if it were possible for a buffer to be revoked and
* discarded, and then reallocated within the same transaction. In such
* a case we would have lost the revoked bit, but when we arrived here
* the second time we would still have a pending revoke to cancel. So,
* do not trust the Revoked bit on buffers unless RevokeValid is also
* set.
*/
int jbd2_journal_cancel_revoke(handle_t *handle, struct journal_head *jh)
{
struct jbd2_revoke_record_s *record;
journal_t *journal = handle->h_transaction->t_journal;
int need_cancel;
int did_revoke = 0; /* akpm: debug */
struct buffer_head *bh = jh2bh(jh);
jbd_debug(4, "journal_head %p, cancelling revoke\n", jh);
/* Is the existing Revoke bit valid? If so, we trust it, and
* only perform the full cancel if the revoke bit is set. If
* not, we can't trust the revoke bit, and we need to do the
* full search for a revoke record. */
if (test_set_buffer_revokevalid(bh)) {
need_cancel = test_clear_buffer_revoked(bh);
} else {
need_cancel = 1;
clear_buffer_revoked(bh);
}
if (need_cancel) {
record = find_revoke_record(journal, bh->b_blocknr);
if (record) {
jbd_debug(4, "cancelled existing revoke on "
"blocknr %llu\n", (unsigned long long)bh->b_blocknr);
spin_lock(&journal->j_revoke_lock);
list_del(&record->hash);
spin_unlock(&journal->j_revoke_lock);
kmem_cache_free(jbd2_revoke_record_cache, record);
did_revoke = 1;
}
}
#ifdef JBD2_EXPENSIVE_CHECKING
/* There better not be one left behind by now! */
record = find_revoke_record(journal, bh->b_blocknr);
J_ASSERT_JH(jh, record == NULL);
#endif
/* Finally, have we just cleared revoke on an unhashed
* buffer_head? If so, we'd better make sure we clear the
* revoked status on any hashed alias too, otherwise the revoke
* state machine will get very upset later on. */
if (need_cancel) {
struct buffer_head *bh2;
bh2 = __find_get_block(bh->b_bdev, bh->b_blocknr, bh->b_size);
if (bh2) {
if (bh2 != bh)
clear_buffer_revoked(bh2);
__brelse(bh2);
}
}
return did_revoke;
}
/*
* journal_clear_revoked_flag clears revoked flag of buffers in
* revoke table to reflect there is no revoked buffers in the next
* transaction which is going to be started.
*/
void jbd2_clear_buffer_revoked_flags(journal_t *journal)
{
struct jbd2_revoke_table_s *revoke = journal->j_revoke;
int i = 0;
for (i = 0; i < revoke->hash_size; i++) {
struct list_head *hash_list;
struct list_head *list_entry;
hash_list = &revoke->hash_table[i];
list_for_each(list_entry, hash_list) {
struct jbd2_revoke_record_s *record;
struct buffer_head *bh;
record = (struct jbd2_revoke_record_s *)list_entry;
bh = __find_get_block(journal->j_fs_dev,
record->blocknr,
journal->j_blocksize);
if (bh) {
clear_buffer_revoked(bh);
__brelse(bh);
}
}
}
}
/* journal_switch_revoke table select j_revoke for next transaction
* we do not want to suspend any processing until all revokes are
* written -bzzz
*/
void jbd2_journal_switch_revoke_table(journal_t *journal)
{
int i;
if (journal->j_revoke == journal->j_revoke_table[0])
journal->j_revoke = journal->j_revoke_table[1];
else
journal->j_revoke = journal->j_revoke_table[0];
for (i = 0; i < journal->j_revoke->hash_size; i++)
INIT_LIST_HEAD(&journal->j_revoke->hash_table[i]);
}
/*
* Write revoke records to the journal for all entries in the current
* revoke hash, deleting the entries as we go.
*/
void jbd2_journal_write_revoke_records(transaction_t *transaction,
struct list_head *log_bufs)
{
journal_t *journal = transaction->t_journal;
struct buffer_head *descriptor;
struct jbd2_revoke_record_s *record;
struct jbd2_revoke_table_s *revoke;
struct list_head *hash_list;
int i, offset, count;
descriptor = NULL;
offset = 0;
count = 0;
/* select revoke table for committing transaction */
revoke = journal->j_revoke == journal->j_revoke_table[0] ?
journal->j_revoke_table[1] : journal->j_revoke_table[0];
for (i = 0; i < revoke->hash_size; i++) {
hash_list = &revoke->hash_table[i];
while (!list_empty(hash_list)) {
record = (struct jbd2_revoke_record_s *)
hash_list->next;
write_one_revoke_record(transaction, log_bufs,
&descriptor, &offset, record);
count++;
list_del(&record->hash);
kmem_cache_free(jbd2_revoke_record_cache, record);
}
}
if (descriptor)
flush_descriptor(journal, descriptor, offset);
jbd_debug(1, "Wrote %d revoke records\n", count);
}
/*
* Write out one revoke record. We need to create a new descriptor
* block if the old one is full or if we have not already created one.
*/
static void write_one_revoke_record(transaction_t *transaction,
struct list_head *log_bufs,
struct buffer_head **descriptorp,
int *offsetp,
struct jbd2_revoke_record_s *record)
{
journal_t *journal = transaction->t_journal;
int csum_size = 0;
struct buffer_head *descriptor;
int sz, offset;
/* If we are already aborting, this all becomes a noop. We
still need to go round the loop in
jbd2_journal_write_revoke_records in order to free all of the
revoke records: only the IO to the journal is omitted. */
if (is_journal_aborted(journal))
return;
descriptor = *descriptorp;
offset = *offsetp;
/* Do we need to leave space at the end for a checksum? */
if (jbd2_journal_has_csum_v2or3(journal))
csum_size = sizeof(struct jbd2_journal_block_tail);
if (jbd2_has_feature_64bit(journal))
sz = 8;
else
sz = 4;
/* Make sure we have a descriptor with space left for the record */
if (descriptor) {
if (offset + sz > journal->j_blocksize - csum_size) {
flush_descriptor(journal, descriptor, offset);
descriptor = NULL;
}
}
if (!descriptor) {
descriptor = jbd2_journal_get_descriptor_buffer(transaction,
JBD2_REVOKE_BLOCK);
if (!descriptor)
return;
/* Record it so that we can wait for IO completion later */
BUFFER_TRACE(descriptor, "file in log_bufs");
jbd2_file_log_bh(log_bufs, descriptor);
offset = sizeof(jbd2_journal_revoke_header_t);
*descriptorp = descriptor;
}
if (jbd2_has_feature_64bit(journal))
* ((__be64 *)(&descriptor->b_data[offset])) =
cpu_to_be64(record->blocknr);
else
* ((__be32 *)(&descriptor->b_data[offset])) =
cpu_to_be32(record->blocknr);
offset += sz;
*offsetp = offset;
}
/*
* Flush a revoke descriptor out to the journal. If we are aborting,
* this is a noop; otherwise we are generating a buffer which needs to
* be waited for during commit, so it has to go onto the appropriate
* journal buffer list.
*/
static void flush_descriptor(journal_t *journal,
struct buffer_head *descriptor,
int offset)
{
jbd2_journal_revoke_header_t *header;
if (is_journal_aborted(journal)) {
put_bh(descriptor);
return;
}
header = (jbd2_journal_revoke_header_t *)descriptor->b_data;
header->r_count = cpu_to_be32(offset);
jbd2_descriptor_block_csum_set(journal, descriptor);
set_buffer_jwrite(descriptor);
BUFFER_TRACE(descriptor, "write");
set_buffer_dirty(descriptor);
write_dirty_buffer(descriptor, REQ_SYNC);
}
#endif
#endif
/*
* Revoke support for recovery.
*
* Recovery needs to be able to:
*
* record all revoke records, including the tid of the latest instance
* of each revoke in the journal
*
* check whether a given block in a given transaction should be replayed
* (ie. has not been revoked by a revoke record in that or a subsequent
* transaction)
*
* empty the revoke table after recovery.
*/
/*
* First, setting revoke records. We create a new revoke record for
* every block ever revoked in the log as we scan it for recovery, and
* we update the existing records if we find multiple revokes for a
* single block.
*/
int jbd2_journal_set_revoke(journal_t *journal,
unsigned long long blocknr,
tid_t sequence)
{
struct jbd2_revoke_record_s *record;
record = find_revoke_record(journal, blocknr);
if (record) {
/* If we have multiple occurrences, only record the
* latest sequence number in the hashed record */
if (tid_gt(sequence, record->sequence))
record->sequence = sequence;
return 0;
}
return insert_revoke_hash(journal, blocknr, sequence);
}
/*
* Test revoke records. For a given block referenced in the log, has
* that block been revoked? A revoke record with a given transaction
* sequence number revokes all blocks in that transaction and earlier
* ones, but later transactions still need replayed.
*/
int jbd2_journal_test_revoke(journal_t *journal,
unsigned long long blocknr,
tid_t sequence)
{
struct jbd2_revoke_record_s *record;
record = find_revoke_record(journal, blocknr);
if (!record)
return 0;
if (tid_gt(sequence, record->sequence))
return 0;
return 1;
}
/*
* Finally, once recovery is over, we need to clear the revoke table so
* that it can be reused by the running filesystem.
*/
void jbd2_journal_clear_revoke(journal_t *journal)
{
int i;
struct list_head *hash_list;
struct jbd2_revoke_record_s *record;
struct jbd2_revoke_table_s *revoke;
revoke = journal->j_revoke;
for (i = 0; i < revoke->hash_size; i++) {
hash_list = &revoke->hash_table[i];
while (!list_empty(hash_list)) {
record = (struct jbd2_revoke_record_s*) hash_list->next;
list_del(&record->hash);
kmem_cache_free(jbd2_revoke_record_cache, record);
}
}
}

54
Ext4Fsd/jbd2/transaction.c Normal file
View File

@@ -0,0 +1,54 @@
// SPDX-License-Identifier: GPL-2.0+
/*
* linux/fs/jbd2/transaction.c
*
* Written by Stephen C. Tweedie <sct@redhat.com>, 1998
*
* Copyright 1998 Red Hat corp --- All Rights Reserved
*
* Generic filesystem transaction handling code; part of the ext2fs
* journaling system.
*
* This file manages transactions (compound commits managed by the
* journaling code) and handles (individual atomic operations by the
* filesystem).
*/
#include <linux/time.h>
#include <linux/fs.h>
#include <linux/jbd2.h>
#include <linux/errno.h>
#include <linux/slab.h>
#include <linux/timer.h>
#include <linux/mm.h>
#include <linux/highmem.h>
//#include <linux/hrtimer.h>
//#include <linux/backing-dev.h>
//#include <linux/bug.h>
#include <linux/module.h>
//#include <linux/sched/mm.h>
//#include <trace/events/jbd2.h>
static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh);
static void __jbd2_journal_unfile_buffer(struct journal_head *jh);
static struct kmem_cache *transaction_cache;
int __init jbd2_journal_init_transaction_cache(void)
{
J_ASSERT(!transaction_cache);
transaction_cache = kmem_cache_create("jbd2_transaction_s",
sizeof(transaction_t),
0,
SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY,
NULL);
if (transaction_cache)
return 0;
return -ENOMEM;
}
void jbd2_journal_destroy_transaction_cache(void)
{
kmem_cache_destroy(transaction_cache);
transaction_cache = NULL;
}

1
Ext4Fsd/linux.c
View File

@@ -10,7 +10,6 @@
/* INCLUDES *****************************************************************/ /* INCLUDES *****************************************************************/
#include <ext2fs.h> #include <ext2fs.h>
#include <linux/jbd.h>
#include <linux/errno.h> #include <linux/errno.h>
/* GLOBALS ***************************************************************/ /* GLOBALS ***************************************************************/