linux-stable/include/linux/jbd.h

1098 lines
33 KiB
C
Raw Normal View History

/*
* linux/include/linux/jbd.h
*
* Written by Stephen C. Tweedie <sct@redhat.com>
*
* Copyright 1998-2000 Red Hat, Inc --- All Rights Reserved
*
* This file is part of the Linux kernel and is made available under
* the terms of the GNU General Public License, version 2, or at your
* option, any later version, incorporated herein by reference.
*
* Definitions for transaction data structures for the buffer cache
* filesystem journaling support.
*/
#ifndef _LINUX_JBD_H
#define _LINUX_JBD_H
/* Allow this file to be included directly into e2fsprogs */
#ifndef __KERNEL__
#include "jfs_compat.h"
#define JFS_DEBUG
#define jfs_debug jbd_debug
#else
[PATCH] jbd: fix transaction batching Ben points out that: When writing files out using O_SYNC, jbd's 1 jiffy delay results in a significant drop in throughput as the disk sits idle. The patch below results in a 4-5x performance improvement (from 6.5MB/s to ~24-30MB/s on my IDE test box) when writing out files using O_SYNC. So optimise the batching code by omitting it entirely if the process which is doing a sync write is the same as the one which did the most recent sync write. If that's true, we're unlikely to get any other processes joining the transaction. (Has been in -mm for ages - it took me a long time to get on to performance testing it) Numbers, on write-cache-disabled IDE: /usr/bin/time -p synctest -n 10 -uf -t 1 -p 1 dir-name Unpatched: 40 seconds Patched: 35 seconds Batching disabled: 35 seconds This is the problematic single-process-doing-fsync case. With multiple fsyncing processes the numbers are AFACIT unaltered by the patch. Aside: performance testing and instrumentation shows that the transaction batching almost doesn't help (testing with synctest -n 1 -uf -t 100 -p 10 dir-name on non-writeback-caching IDE). This is because by the time one process is running a synchronous commit, a bunch of other processes already have a transaction handle open, so they're all going to batch into the same transaction anyway. The batching seems to offer maybe 5-10% speedup with this workload, but I'm pretty sure it was more important than that when it was first developed 4-odd years ago... Cc: "Stephen C. Tweedie" <sct@redhat.com> Cc: Benjamin LaHaise <bcrl@kvack.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-05 07:27:54 +00:00
#include <linux/types.h>
#include <linux/buffer_head.h>
#include <linux/journal-head.h>
#include <linux/stddef.h>
[PATCH] spinlock consolidation This patch (written by me and also containing many suggestions of Arjan van de Ven) does a major cleanup of the spinlock code. It does the following things: - consolidates and enhances the spinlock/rwlock debugging code - simplifies the asm/spinlock.h files - encapsulates the raw spinlock type and moves generic spinlock features (such as ->break_lock) into the generic code. - cleans up the spinlock code hierarchy to get rid of the spaghetti. Most notably there's now only a single variant of the debugging code, located in lib/spinlock_debug.c. (previously we had one SMP debugging variant per architecture, plus a separate generic one for UP builds) Also, i've enhanced the rwlock debugging facility, it will now track write-owners. There is new spinlock-owner/CPU-tracking on SMP builds too. All locks have lockup detection now, which will work for both soft and hard spin/rwlock lockups. The arch-level include files now only contain the minimally necessary subset of the spinlock code - all the rest that can be generalized now lives in the generic headers: include/asm-i386/spinlock_types.h | 16 include/asm-x86_64/spinlock_types.h | 16 I have also split up the various spinlock variants into separate files, making it easier to see which does what. The new layout is: SMP | UP ----------------------------|----------------------------------- asm/spinlock_types_smp.h | linux/spinlock_types_up.h linux/spinlock_types.h | linux/spinlock_types.h asm/spinlock_smp.h | linux/spinlock_up.h linux/spinlock_api_smp.h | linux/spinlock_api_up.h linux/spinlock.h | linux/spinlock.h /* * here's the role of the various spinlock/rwlock related include files: * * on SMP builds: * * asm/spinlock_types.h: contains the raw_spinlock_t/raw_rwlock_t and the * initializers * * linux/spinlock_types.h: * defines the generic type and initializers * * asm/spinlock.h: contains the __raw_spin_*()/etc. lowlevel * implementations, mostly inline assembly code * * (also included on UP-debug builds:) * * linux/spinlock_api_smp.h: * contains the prototypes for the _spin_*() APIs. * * linux/spinlock.h: builds the final spin_*() APIs. * * on UP builds: * * linux/spinlock_type_up.h: * contains the generic, simplified UP spinlock type. * (which is an empty structure on non-debug builds) * * linux/spinlock_types.h: * defines the generic type and initializers * * linux/spinlock_up.h: * contains the __raw_spin_*()/etc. version of UP * builds. (which are NOPs on non-debug, non-preempt * builds) * * (included on UP-non-debug builds:) * * linux/spinlock_api_up.h: * builds the _spin_*() APIs. * * linux/spinlock.h: builds the final spin_*() APIs. */ All SMP and UP architectures are converted by this patch. arm, i386, ia64, ppc, ppc64, s390/s390x, x64 was build-tested via crosscompilers. m32r, mips, sh, sparc, have not been tested yet, but should be mostly fine. From: Grant Grundler <grundler@parisc-linux.org> Booted and lightly tested on a500-44 (64-bit, SMP kernel, dual CPU). Builds 32-bit SMP kernel (not booted or tested). I did not try to build non-SMP kernels. That should be trivial to fix up later if necessary. I converted bit ops atomic_hash lock to raw_spinlock_t. Doing so avoids some ugly nesting of linux/*.h and asm/*.h files. Those particular locks are well tested and contained entirely inside arch specific code. I do NOT expect any new issues to arise with them. If someone does ever need to use debug/metrics with them, then they will need to unravel this hairball between spinlocks, atomic ops, and bit ops that exist only because parisc has exactly one atomic instruction: LDCW (load and clear word). From: "Luck, Tony" <tony.luck@intel.com> ia64 fix Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Arjan van de Ven <arjanv@infradead.org> Signed-off-by: Grant Grundler <grundler@parisc-linux.org> Cc: Matthew Wilcox <willy@debian.org> Signed-off-by: Hirokazu Takata <takata@linux-m32r.org> Signed-off-by: Mikael Pettersson <mikpe@csd.uu.se> Signed-off-by: Benoit Boissinot <benoit.boissinot@ens-lyon.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-09-10 07:25:56 +00:00
#include <linux/bit_spinlock.h>
#include <linux/mutex.h>
#include <linux/timer.h>
#include <linux/lockdep.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
#define journal_oom_retry 1
/*
* Define JBD_PARANOID_IOFAIL to cause a kernel BUG() if ext3 finds
* certain classes of error which can occur due to failed IOs. Under
* normal use we want ext3 to continue after such errors, because
* hardware _can_ fail, but for debugging purposes when running tests on
* known-good hardware we may want to trap these errors.
*/
#undef JBD_PARANOID_IOFAIL
/*
* The default maximum commit age, in seconds.
*/
#define JBD_DEFAULT_MAX_COMMIT_AGE 5
#ifdef CONFIG_JBD_DEBUG
/*
* Define JBD_EXPENSIVE_CHECKING to enable more expensive internal
* consistency checks. By default we don't do this unless
* CONFIG_JBD_DEBUG is on.
*/
#define JBD_EXPENSIVE_CHECKING
extern u8 journal_enable_debug;
#define jbd_debug(n, f, a...) \
do { \
if ((n) <= journal_enable_debug) { \
printk (KERN_DEBUG "(%s, %d): %s: ", \
__FILE__, __LINE__, __func__); \
printk (f, ## a); \
} \
} while (0)
#else
#define jbd_debug(f, a...) /**/
#endif
static inline void *jbd_alloc(size_t size, gfp_t flags)
{
return (void *)__get_free_pages(flags, get_order(size));
}
static inline void jbd_free(void *ptr, size_t size)
{
free_pages((unsigned long)ptr, get_order(size));
};
#define JFS_MIN_JOURNAL_BLOCKS 1024
/**
* typedef handle_t - The handle_t type represents a single atomic update being performed by some process.
*
* All filesystem modifications made by the process go
* through this handle. Recursive operations (such as quota operations)
* are gathered into a single update.
*
* The buffer credits field is used to account for journaled buffers
* being modified by the running process. To ensure that there is
* enough log space for all outstanding operations, we need to limit the
* number of outstanding buffers possible at any time. When the
* operation completes, any buffer credits not used are credited back to
* the transaction, so that at all times we know how many buffers the
* outstanding updates on a transaction might possibly touch.
*
* This is an opaque datatype.
**/
typedef struct handle_s handle_t; /* Atomic operation type */
/**
* typedef journal_t - The journal_t maintains all of the journaling state information for a single filesystem.
*
* journal_t is linked to from the fs superblock structure.
*
* We use the journal_t to keep track of all outstanding transaction
* activity on the filesystem, and to manage the state of the log
* writing process.
*
* This is an opaque datatype.
**/
typedef struct journal_s journal_t; /* Journal control structure */
#endif
/*
* Internal structures used by the logging mechanism:
*/
#define JFS_MAGIC_NUMBER 0xc03b3998U /* The first 4 bytes of /dev/random! */
/*
* On-disk structures
*/
/*
* Descriptor block types:
*/
#define JFS_DESCRIPTOR_BLOCK 1
#define JFS_COMMIT_BLOCK 2
#define JFS_SUPERBLOCK_V1 3
#define JFS_SUPERBLOCK_V2 4
#define JFS_REVOKE_BLOCK 5
/*
* Standard header for all descriptor blocks:
*/
typedef struct journal_header_s
{
__be32 h_magic;
__be32 h_blocktype;
__be32 h_sequence;
} journal_header_t;
/*
* The block tag: used to describe a single buffer in the journal
*/
typedef struct journal_block_tag_s
{
__be32 t_blocknr; /* The on-disk block number */
__be32 t_flags; /* See below */
} journal_block_tag_t;
/*
* The revoke descriptor: used on disk to describe a series of blocks to
* be revoked from the log
*/
typedef struct journal_revoke_header_s
{
journal_header_t r_header;
__be32 r_count; /* Count of bytes used in the block */
} journal_revoke_header_t;
/* Definitions for the journal tag flags word: */
#define JFS_FLAG_ESCAPE 1 /* on-disk block is escaped */
#define JFS_FLAG_SAME_UUID 2 /* block has same uuid as previous */
#define JFS_FLAG_DELETED 4 /* block deleted by this transaction */
#define JFS_FLAG_LAST_TAG 8 /* last tag in this descriptor block */
/*
* The journal superblock. All fields are in big-endian byte order.
*/
typedef struct journal_superblock_s
{
/* 0x0000 */
journal_header_t s_header;
/* 0x000C */
/* Static information describing the journal */
__be32 s_blocksize; /* journal device blocksize */
__be32 s_maxlen; /* total blocks in journal file */
__be32 s_first; /* first block of log information */
/* 0x0018 */
/* Dynamic information describing the current state of the log */
__be32 s_sequence; /* first commit ID expected in log */
__be32 s_start; /* blocknr of start of log */
/* 0x0020 */
/* Error value, as set by journal_abort(). */
__be32 s_errno;
/* 0x0024 */
/* Remaining fields are only valid in a version-2 superblock */
__be32 s_feature_compat; /* compatible feature set */
__be32 s_feature_incompat; /* incompatible feature set */
__be32 s_feature_ro_compat; /* readonly-compatible feature set */
/* 0x0030 */
__u8 s_uuid[16]; /* 128-bit uuid for journal */
/* 0x0040 */
__be32 s_nr_users; /* Nr of filesystems sharing log */
__be32 s_dynsuper; /* Blocknr of dynamic superblock copy*/
/* 0x0048 */
__be32 s_max_transaction; /* Limit of journal blocks per trans.*/
__be32 s_max_trans_data; /* Limit of data blocks per trans. */
/* 0x0050 */
__u32 s_padding[44];
/* 0x0100 */
__u8 s_users[16*48]; /* ids of all fs'es sharing the log */
/* 0x0400 */
} journal_superblock_t;
#define JFS_HAS_COMPAT_FEATURE(j,mask) \
((j)->j_format_version >= 2 && \
((j)->j_superblock->s_feature_compat & cpu_to_be32((mask))))
#define JFS_HAS_RO_COMPAT_FEATURE(j,mask) \
((j)->j_format_version >= 2 && \
((j)->j_superblock->s_feature_ro_compat & cpu_to_be32((mask))))
#define JFS_HAS_INCOMPAT_FEATURE(j,mask) \
((j)->j_format_version >= 2 && \
((j)->j_superblock->s_feature_incompat & cpu_to_be32((mask))))
#define JFS_FEATURE_INCOMPAT_REVOKE 0x00000001
/* Features known to this kernel version: */
#define JFS_KNOWN_COMPAT_FEATURES 0
#define JFS_KNOWN_ROCOMPAT_FEATURES 0
#define JFS_KNOWN_INCOMPAT_FEATURES JFS_FEATURE_INCOMPAT_REVOKE
#ifdef __KERNEL__
#include <linux/fs.h>
#include <linux/sched.h>
#define J_ASSERT(assert) BUG_ON(!(assert))
#define J_ASSERT_BH(bh, expr) J_ASSERT(expr)
#define J_ASSERT_JH(jh, expr) J_ASSERT(expr)
#if defined(JBD_PARANOID_IOFAIL)
#define J_EXPECT(expr, why...) J_ASSERT(expr)
#define J_EXPECT_BH(bh, expr, why...) J_ASSERT_BH(bh, expr)
#define J_EXPECT_JH(jh, expr, why...) J_ASSERT_JH(jh, expr)
#else
#define __journal_expect(expr, why...) \
({ \
int val = (expr); \
if (!val) { \
printk(KERN_ERR \
"EXT3-fs unexpected failure: %s;\n",# expr); \
printk(KERN_ERR why "\n"); \
} \
val; \
})
#define J_EXPECT(expr, why...) __journal_expect(expr, ## why)
#define J_EXPECT_BH(bh, expr, why...) __journal_expect(expr, ## why)
#define J_EXPECT_JH(jh, expr, why...) __journal_expect(expr, ## why)
#endif
enum jbd_state_bits {
BH_JBD /* Has an attached ext3 journal_head */
= BH_PrivateStart,
BH_JWrite, /* Being written to log (@@@ DEBUGGING) */
BH_Freed, /* Has been freed (truncated) */
BH_Revoked, /* Has been revoked from the log */
BH_RevokeValid, /* Revoked flag is valid */
BH_JBDDirty, /* Is dirty but journaled */
BH_State, /* Pins most journal_head state */
BH_JournalHead, /* Pins bh->b_private and jh->b_bh */
BH_Unshadow, /* Dummy bit, for BJ_Shadow wakeup filtering */
};
BUFFER_FNS(JBD, jbd)
BUFFER_FNS(JWrite, jwrite)
BUFFER_FNS(JBDDirty, jbddirty)
TAS_BUFFER_FNS(JBDDirty, jbddirty)
BUFFER_FNS(Revoked, revoked)
TAS_BUFFER_FNS(Revoked, revoked)
BUFFER_FNS(RevokeValid, revokevalid)
TAS_BUFFER_FNS(RevokeValid, revokevalid)
BUFFER_FNS(Freed, freed)
static inline struct buffer_head *jh2bh(struct journal_head *jh)
{
return jh->b_bh;
}
static inline struct journal_head *bh2jh(struct buffer_head *bh)
{
return bh->b_private;
}
static inline void jbd_lock_bh_state(struct buffer_head *bh)
{
bit_spin_lock(BH_State, &bh->b_state);
}
static inline int jbd_trylock_bh_state(struct buffer_head *bh)
{
return bit_spin_trylock(BH_State, &bh->b_state);
}
static inline int jbd_is_locked_bh_state(struct buffer_head *bh)
{
return bit_spin_is_locked(BH_State, &bh->b_state);
}
static inline void jbd_unlock_bh_state(struct buffer_head *bh)
{
bit_spin_unlock(BH_State, &bh->b_state);
}
static inline void jbd_lock_bh_journal_head(struct buffer_head *bh)
{
bit_spin_lock(BH_JournalHead, &bh->b_state);
}
static inline void jbd_unlock_bh_journal_head(struct buffer_head *bh)
{
bit_spin_unlock(BH_JournalHead, &bh->b_state);
}
struct jbd_revoke_table_s;
/**
fs: fix kernel-doc notation warnings Fix kernel-doc notation warnings in fs/. Warning(mmotm-2008-0314-1449//fs/super.c:560): missing initial short description on line: * mark_files_ro Warning(mmotm-2008-0314-1449//fs/locks.c:1277): missing initial short description on line: * lease_get_mtime Warning(mmotm-2008-0314-1449//fs/locks.c:1277): missing initial short description on line: * lease_get_mtime Warning(mmotm-2008-0314-1449//fs/namei.c:1368): missing initial short description on line: * lookup_one_len: filesystem helper to lookup single pathname component Warning(mmotm-2008-0314-1449//fs/buffer.c:3221): missing initial short description on line: * bh_uptodate_or_lock: Test whether the buffer is uptodate Warning(mmotm-2008-0314-1449//fs/buffer.c:3240): missing initial short description on line: * bh_submit_read: Submit a locked buffer for reading Warning(mmotm-2008-0314-1449//fs/fs-writeback.c:30): missing initial short description on line: * writeback_acquire: attempt to get exclusive writeback access to a device Warning(mmotm-2008-0314-1449//fs/fs-writeback.c:47): missing initial short description on line: * writeback_in_progress: determine whether there is writeback in progress Warning(mmotm-2008-0314-1449//fs/fs-writeback.c:58): missing initial short description on line: * writeback_release: relinquish exclusive writeback access against a device. Warning(mmotm-2008-0314-1449//include/linux/jbd.h:351): contents before sections Warning(mmotm-2008-0314-1449//include/linux/jbd.h:561): contents before sections Warning(mmotm-2008-0314-1449//fs/jbd/transaction.c:1935): missing initial short description on line: * void journal_invalidatepage() Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-03-20 00:01:00 +00:00
* struct handle_s - this is the concrete type associated with handle_t.
* @h_transaction: Which compound transaction is this update a part of?
* @h_buffer_credits: Number of remaining buffers we are allowed to dirty.
* @h_ref: Reference count on this handle
* @h_err: Field for caller's use to track errors through large fs operations
* @h_sync: flag for sync-on-close
* @h_jdata: flag to force data journaling
* @h_aborted: flag indicating fatal error on handle
* @h_lockdep_map: lockdep info for debugging lock problems
*/
struct handle_s
{
/* Which compound transaction is this update a part of? */
transaction_t *h_transaction;
/* Number of remaining buffers we are allowed to dirty: */
int h_buffer_credits;
/* Reference count on this handle */
int h_ref;
/* Field for caller's use to track errors through large fs */
/* operations */
int h_err;
/* Flags [no locking] */
unsigned int h_sync: 1; /* sync-on-close */
unsigned int h_jdata: 1; /* force data journaling */
unsigned int h_aborted: 1; /* fatal error on handle */
#ifdef CONFIG_DEBUG_LOCK_ALLOC
struct lockdep_map h_lockdep_map;
#endif
};
/* The transaction_t type is the guts of the journaling mechanism. It
* tracks a compound transaction through its various states:
*
* RUNNING: accepting new updates
* LOCKED: Updates still running but we don't accept new ones
* RUNDOWN: Updates are tidying up but have finished requesting
* new buffers to modify (state not used for now)
* FLUSH: All updates complete, but we are still writing to disk
* COMMIT: All data on disk, writing commit record
* FINISHED: We still have to keep the transaction for checkpointing.
*
* The transaction keeps track of all of the buffers modified by a
* running transaction, and all of the buffers committed but not yet
* flushed to home for finished transactions.
*/
/*
* Lock ranking:
*
* j_list_lock
* ->jbd_lock_bh_journal_head() (This is "innermost")
*
* j_state_lock
* ->jbd_lock_bh_state()
*
* jbd_lock_bh_state()
* ->j_list_lock
*
* j_state_lock
* ->t_handle_lock
*
* j_state_lock
* ->j_list_lock (journal_unmap_buffer)
*
*/
struct transaction_s
{
/* Pointer to the journal for this transaction. [no locking] */
journal_t *t_journal;
/* Sequence number for this transaction [no locking] */
tid_t t_tid;
/*
* Transaction's current state
* [no locking - only kjournald alters this]
* [j_list_lock] guards transition of a transaction into T_FINISHED
* state and subsequent call of __journal_drop_transaction()
* FIXME: needs barriers
* KLUDGE: [use j_state_lock]
*/
enum {
T_RUNNING,
T_LOCKED,
T_FLUSH,
T_COMMIT,
T_COMMIT_RECORD,
T_FINISHED
} t_state;
/*
* Where in the log does this transaction's commit start? [no locking]
*/
unsigned int t_log_start;
/* Number of buffers on the t_buffers list [j_list_lock] */
int t_nr_buffers;
/*
* Doubly-linked circular list of all buffers reserved but not yet
* modified by this transaction [j_list_lock]
*/
struct journal_head *t_reserved_list;
/*
* Doubly-linked circular list of all buffers under writeout during
* commit [j_list_lock]
*/
struct journal_head *t_locked_list;
/*
* Doubly-linked circular list of all metadata buffers owned by this
* transaction [j_list_lock]
*/
struct journal_head *t_buffers;
/*
* Doubly-linked circular list of all data buffers still to be
* flushed before this transaction can be committed [j_list_lock]
*/
struct journal_head *t_sync_datalist;
/*
* Doubly-linked circular list of all forget buffers (superseded
* buffers which we can un-checkpoint once this transaction commits)
* [j_list_lock]
*/
struct journal_head *t_forget;
/*
* Doubly-linked circular list of all buffers still to be flushed before
* this transaction can be checkpointed. [j_list_lock]
*/
struct journal_head *t_checkpoint_list;
/*
* Doubly-linked circular list of all buffers submitted for IO while
* checkpointing. [j_list_lock]
*/
struct journal_head *t_checkpoint_io_list;
/*
* Doubly-linked circular list of temporary buffers currently undergoing
* IO in the log [j_list_lock]
*/
struct journal_head *t_iobuf_list;
/*
* Doubly-linked circular list of metadata buffers being shadowed by log
* IO. The IO buffers on the iobuf list and the shadow buffers on this
* list match each other one for one at all times. [j_list_lock]
*/
struct journal_head *t_shadow_list;
/*
* Doubly-linked circular list of control buffers being written to the
* log. [j_list_lock]
*/
struct journal_head *t_log_list;
/*
* Protects info related to handles
*/
spinlock_t t_handle_lock;
/*
* Number of outstanding updates running on this transaction
* [t_handle_lock]
*/
int t_updates;
/*
* Number of buffers reserved for use by all handles in this transaction
* handle but not yet modified. [t_handle_lock]
*/
int t_outstanding_credits;
/*
* Forward and backward links for the circular list of all transactions
* awaiting checkpoint. [j_list_lock]
*/
transaction_t *t_cpnext, *t_cpprev;
/*
* When will the transaction expire (become due for commit), in jiffies?
* [no locking]
*/
unsigned long t_expires;
jbd: improve fsync batching There is a flaw with the way jbd handles fsync batching. If we fsync() a file and we were not the last person to run fsync() on this fs then we automatically sleep for 1 jiffie in order to wait for new writers to join into the transaction before forcing the commit. The problem with this is that with really fast storage (ie a Clariion) the time it takes to commit a transaction to disk is way faster than 1 jiffie in most cases, so sleeping means waiting longer with nothing to do than if we just committed the transaction and kept going. Ric Wheeler noticed this when using fs_mark with more than 1 thread, the throughput would plummet as he added more threads. This patch attempts to fix this problem by recording the average time in nanoseconds that it takes to commit a transaction to disk, and what time we started the transaction. If we run an fsync() and we have been running for less time than it takes to commit the transaction to disk, we sleep for the delta amount of time and then commit to disk. We acheive sub-jiffie sleeping using schedule_hrtimeout. This means that the wait time is auto-tuned to the speed of the underlying disk, instead of having this static timeout. I weighted the average according to somebody's comments (Andreas Dilger I think) in order to help normalize random outliers where we take way longer or way less time to commit than the average. I also have a min() check in there to make sure we don't sleep longer than a jiffie in case our storage is super slow, this was requested by Andrew. I unfortunately do not have access to a Clariion, so I had to use a ramdisk to represent a super fast array. I tested with a SATA drive with barrier=1 to make sure there was no regression with local disks, I tested with a 4 way multipathed Apple Xserve RAID array and of course the ramdisk. I ran the following command fs_mark -d /mnt/ext3-test -s 4096 -n 2000 -D 64 -t $i where $i was 2, 4, 8, 16 and 32. I mkfs'ed the fs each time. Here are my results type threads with patch without patch sata 2 24.6 26.3 sata 4 49.2 48.1 sata 8 70.1 67.0 sata 16 104.0 94.1 sata 32 153.6 142.7 xserve 2 246.4 222.0 xserve 4 480.0 440.8 xserve 8 829.5 730.8 xserve 16 1172.7 1026.9 xserve 32 1816.3 1650.5 ramdisk 2 2538.3 1745.6 ramdisk 4 2942.3 661.9 ramdisk 8 2882.5 999.8 ramdisk 16 2738.7 1801.9 ramdisk 32 2541.9 2394.0 Signed-off-by: Josef Bacik <jbacik@redhat.com> Cc: Andreas Dilger <adilger@sun.com> Cc: Arjan van de Ven <arjan@infradead.org> Cc: Ric Wheeler <rwheeler@redhat.com> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-08 02:07:24 +00:00
/*
* When this transaction started, in nanoseconds [no locking]
*/
ktime_t t_start_time;
/*
* How many handles used this transaction? [t_handle_lock]
*/
int t_handle_count;
/*
* This transaction is being forced and some process is
* waiting for it to finish.
*/
unsigned int t_synchronous_commit:1;
};
/**
fs: fix kernel-doc notation warnings Fix kernel-doc notation warnings in fs/. Warning(mmotm-2008-0314-1449//fs/super.c:560): missing initial short description on line: * mark_files_ro Warning(mmotm-2008-0314-1449//fs/locks.c:1277): missing initial short description on line: * lease_get_mtime Warning(mmotm-2008-0314-1449//fs/locks.c:1277): missing initial short description on line: * lease_get_mtime Warning(mmotm-2008-0314-1449//fs/namei.c:1368): missing initial short description on line: * lookup_one_len: filesystem helper to lookup single pathname component Warning(mmotm-2008-0314-1449//fs/buffer.c:3221): missing initial short description on line: * bh_uptodate_or_lock: Test whether the buffer is uptodate Warning(mmotm-2008-0314-1449//fs/buffer.c:3240): missing initial short description on line: * bh_submit_read: Submit a locked buffer for reading Warning(mmotm-2008-0314-1449//fs/fs-writeback.c:30): missing initial short description on line: * writeback_acquire: attempt to get exclusive writeback access to a device Warning(mmotm-2008-0314-1449//fs/fs-writeback.c:47): missing initial short description on line: * writeback_in_progress: determine whether there is writeback in progress Warning(mmotm-2008-0314-1449//fs/fs-writeback.c:58): missing initial short description on line: * writeback_release: relinquish exclusive writeback access against a device. Warning(mmotm-2008-0314-1449//include/linux/jbd.h:351): contents before sections Warning(mmotm-2008-0314-1449//include/linux/jbd.h:561): contents before sections Warning(mmotm-2008-0314-1449//fs/jbd/transaction.c:1935): missing initial short description on line: * void journal_invalidatepage() Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-03-20 00:01:00 +00:00
* struct journal_s - this is the concrete type associated with journal_t.
* @j_flags: General journaling state flags
* @j_errno: Is there an outstanding uncleared error on the journal (from a
* prior abort)?
* @j_sb_buffer: First part of superblock buffer
* @j_superblock: Second part of superblock buffer
* @j_format_version: Version of the superblock format
* @j_state_lock: Protect the various scalars in the journal
* @j_barrier_count: Number of processes waiting to create a barrier lock
* @j_barrier: The barrier lock itself
* @j_running_transaction: The current running transaction..
* @j_committing_transaction: the transaction we are pushing to disk
* @j_checkpoint_transactions: a linked circular list of all transactions
* waiting for checkpointing
* @j_wait_transaction_locked: Wait queue for waiting for a locked transaction
* to start committing, or for a barrier lock to be released
* @j_wait_logspace: Wait queue for waiting for checkpointing to complete
* @j_wait_done_commit: Wait queue for waiting for commit to complete
* @j_wait_checkpoint: Wait queue to trigger checkpointing
* @j_wait_commit: Wait queue to trigger commit
* @j_wait_updates: Wait queue to wait for updates to complete
* @j_checkpoint_mutex: Mutex for locking against concurrent checkpoints
* @j_head: Journal head - identifies the first unused block in the journal
* @j_tail: Journal tail - identifies the oldest still-used block in the
* journal.
* @j_free: Journal free - how many free blocks are there in the journal?
* @j_first: The block number of the first usable block
* @j_last: The block number one beyond the last usable block
* @j_dev: Device where we store the journal
* @j_blocksize: blocksize for the location where we store the journal.
* @j_blk_offset: starting block offset for into the device where we store the
* journal
* @j_fs_dev: Device which holds the client fs. For internal journal this will
* be equal to j_dev
* @j_maxlen: Total maximum capacity of the journal region on disk.
* @j_list_lock: Protects the buffer lists and internal buffer state.
* @j_inode: Optional inode where we store the journal. If present, all journal
* block numbers are mapped into this inode via bmap().
* @j_tail_sequence: Sequence number of the oldest transaction in the log
* @j_transaction_sequence: Sequence number of the next transaction to grant
* @j_commit_sequence: Sequence number of the most recently committed
* transaction
* @j_commit_request: Sequence number of the most recent transaction wanting
* commit
* @j_uuid: Uuid of client object.
* @j_task: Pointer to the current commit thread for this journal
* @j_max_transaction_buffers: Maximum number of metadata buffers to allow in a
* single compound commit transaction
* @j_commit_interval: What is the maximum transaction lifetime before we begin
* a commit?
* @j_commit_timer: The timer used to wakeup the commit thread
* @j_revoke_lock: Protect the revoke table
* @j_revoke: The revoke table - maintains the list of revoked blocks in the
* current transaction.
* @j_revoke_table: alternate revoke tables for j_revoke
* @j_wbuf: array of buffer_heads for journal_commit_transaction
* @j_wbufsize: maximum number of buffer_heads allowed in j_wbuf, the
* number that will fit in j_blocksize
[PATCH] jbd: fix transaction batching Ben points out that: When writing files out using O_SYNC, jbd's 1 jiffy delay results in a significant drop in throughput as the disk sits idle. The patch below results in a 4-5x performance improvement (from 6.5MB/s to ~24-30MB/s on my IDE test box) when writing out files using O_SYNC. So optimise the batching code by omitting it entirely if the process which is doing a sync write is the same as the one which did the most recent sync write. If that's true, we're unlikely to get any other processes joining the transaction. (Has been in -mm for ages - it took me a long time to get on to performance testing it) Numbers, on write-cache-disabled IDE: /usr/bin/time -p synctest -n 10 -uf -t 1 -p 1 dir-name Unpatched: 40 seconds Patched: 35 seconds Batching disabled: 35 seconds This is the problematic single-process-doing-fsync case. With multiple fsyncing processes the numbers are AFACIT unaltered by the patch. Aside: performance testing and instrumentation shows that the transaction batching almost doesn't help (testing with synctest -n 1 -uf -t 100 -p 10 dir-name on non-writeback-caching IDE). This is because by the time one process is running a synchronous commit, a bunch of other processes already have a transaction handle open, so they're all going to batch into the same transaction anyway. The batching seems to offer maybe 5-10% speedup with this workload, but I'm pretty sure it was more important than that when it was first developed 4-odd years ago... Cc: "Stephen C. Tweedie" <sct@redhat.com> Cc: Benjamin LaHaise <bcrl@kvack.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-05 07:27:54 +00:00
* @j_last_sync_writer: most recent pid which did a synchronous write
* @j_average_commit_time: the average amount of time in nanoseconds it
* takes to commit a transaction to the disk.
* @j_private: An opaque pointer to fs-private information.
*/
struct journal_s
{
/* General journaling state flags [j_state_lock] */
unsigned long j_flags;
/*
* Is there an outstanding uncleared error on the journal (from a prior
* abort)? [j_state_lock]
*/
int j_errno;
/* The superblock buffer */
struct buffer_head *j_sb_buffer;
journal_superblock_t *j_superblock;
/* Version of the superblock format */
int j_format_version;
/*
* Protect the various scalars in the journal
*/
spinlock_t j_state_lock;
/*
* Number of processes waiting to create a barrier lock [j_state_lock]
*/
int j_barrier_count;
/* The barrier lock itself */
struct mutex j_barrier;
/*
* Transactions: The current running transaction...
* [j_state_lock] [caller holding open handle]
*/
transaction_t *j_running_transaction;
/*
* the transaction we are pushing to disk
* [j_state_lock] [caller holding open handle]
*/
transaction_t *j_committing_transaction;
/*
* ... and a linked circular list of all transactions waiting for
* checkpointing. [j_list_lock]
*/
transaction_t *j_checkpoint_transactions;
/*
* Wait queue for waiting for a locked transaction to start committing,
* or for a barrier lock to be released
*/
wait_queue_head_t j_wait_transaction_locked;
/* Wait queue for waiting for checkpointing to complete */
wait_queue_head_t j_wait_logspace;
/* Wait queue for waiting for commit to complete */
wait_queue_head_t j_wait_done_commit;
/* Wait queue to trigger checkpointing */
wait_queue_head_t j_wait_checkpoint;
/* Wait queue to trigger commit */
wait_queue_head_t j_wait_commit;
/* Wait queue to wait for updates to complete */
wait_queue_head_t j_wait_updates;
/* Semaphore for locking against concurrent checkpoints */
struct mutex j_checkpoint_mutex;
/*
* Journal head: identifies the first unused block in the journal.
* [j_state_lock]
*/
unsigned int j_head;
/*
* Journal tail: identifies the oldest still-used block in the journal.
* [j_state_lock]
*/
unsigned int j_tail;
/*
* Journal free: how many free blocks are there in the journal?
* [j_state_lock]
*/
unsigned int j_free;
/*
* Journal start and end: the block numbers of the first usable block
* and one beyond the last usable block in the journal. [j_state_lock]
*/
unsigned int j_first;
unsigned int j_last;
/*
* Device, blocksize and starting block offset for the location where we
* store the journal.
*/
struct block_device *j_dev;
int j_blocksize;
unsigned int j_blk_offset;
/*
* Device which holds the client fs. For internal journal this will be
* equal to j_dev.
*/
struct block_device *j_fs_dev;
/* Total maximum capacity of the journal region on disk. */
unsigned int j_maxlen;
/*
* Protects the buffer lists and internal buffer state.
*/
spinlock_t j_list_lock;
/* Optional inode where we store the journal. If present, all */
/* journal block numbers are mapped into this inode via */
/* bmap(). */
struct inode *j_inode;
/*
* Sequence number of the oldest transaction in the log [j_state_lock]
*/
tid_t j_tail_sequence;
/*
* Sequence number of the next transaction to grant [j_state_lock]
*/
tid_t j_transaction_sequence;
/*
* Sequence number of the most recently committed transaction
* [j_state_lock].
*/
tid_t j_commit_sequence;
/*
* Sequence number of the most recent transaction wanting commit
* [j_state_lock]
*/
tid_t j_commit_request;
/*
* Journal uuid: identifies the object (filesystem, LVM volume etc)
* backed by this journal. This will eventually be replaced by an array
* of uuids, allowing us to index multiple devices within a single
* journal and to perform atomic updates across them.
*/
__u8 j_uuid[16];
/* Pointer to the current commit thread for this journal */
struct task_struct *j_task;
/*
* Maximum number of metadata buffers to allow in a single compound
* commit transaction
*/
int j_max_transaction_buffers;
/*
* What is the maximum transaction lifetime before we begin a commit?
*/
unsigned long j_commit_interval;
/* The timer used to wakeup the commit thread: */
struct timer_list j_commit_timer;
/*
* The revoke table: maintains the list of revoked blocks in the
* current transaction. [j_revoke_lock]
*/
spinlock_t j_revoke_lock;
struct jbd_revoke_table_s *j_revoke;
struct jbd_revoke_table_s *j_revoke_table[2];
/*
* array of bhs for journal_commit_transaction
*/
struct buffer_head **j_wbuf;
int j_wbufsize;
jbd: improve fsync batching There is a flaw with the way jbd handles fsync batching. If we fsync() a file and we were not the last person to run fsync() on this fs then we automatically sleep for 1 jiffie in order to wait for new writers to join into the transaction before forcing the commit. The problem with this is that with really fast storage (ie a Clariion) the time it takes to commit a transaction to disk is way faster than 1 jiffie in most cases, so sleeping means waiting longer with nothing to do than if we just committed the transaction and kept going. Ric Wheeler noticed this when using fs_mark with more than 1 thread, the throughput would plummet as he added more threads. This patch attempts to fix this problem by recording the average time in nanoseconds that it takes to commit a transaction to disk, and what time we started the transaction. If we run an fsync() and we have been running for less time than it takes to commit the transaction to disk, we sleep for the delta amount of time and then commit to disk. We acheive sub-jiffie sleeping using schedule_hrtimeout. This means that the wait time is auto-tuned to the speed of the underlying disk, instead of having this static timeout. I weighted the average according to somebody's comments (Andreas Dilger I think) in order to help normalize random outliers where we take way longer or way less time to commit than the average. I also have a min() check in there to make sure we don't sleep longer than a jiffie in case our storage is super slow, this was requested by Andrew. I unfortunately do not have access to a Clariion, so I had to use a ramdisk to represent a super fast array. I tested with a SATA drive with barrier=1 to make sure there was no regression with local disks, I tested with a 4 way multipathed Apple Xserve RAID array and of course the ramdisk. I ran the following command fs_mark -d /mnt/ext3-test -s 4096 -n 2000 -D 64 -t $i where $i was 2, 4, 8, 16 and 32. I mkfs'ed the fs each time. Here are my results type threads with patch without patch sata 2 24.6 26.3 sata 4 49.2 48.1 sata 8 70.1 67.0 sata 16 104.0 94.1 sata 32 153.6 142.7 xserve 2 246.4 222.0 xserve 4 480.0 440.8 xserve 8 829.5 730.8 xserve 16 1172.7 1026.9 xserve 32 1816.3 1650.5 ramdisk 2 2538.3 1745.6 ramdisk 4 2942.3 661.9 ramdisk 8 2882.5 999.8 ramdisk 16 2738.7 1801.9 ramdisk 32 2541.9 2394.0 Signed-off-by: Josef Bacik <jbacik@redhat.com> Cc: Andreas Dilger <adilger@sun.com> Cc: Arjan van de Ven <arjan@infradead.org> Cc: Ric Wheeler <rwheeler@redhat.com> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-08 02:07:24 +00:00
/*
* this is the pid of the last person to run a synchronous operation
* through the journal.
*/
[PATCH] jbd: fix transaction batching Ben points out that: When writing files out using O_SYNC, jbd's 1 jiffy delay results in a significant drop in throughput as the disk sits idle. The patch below results in a 4-5x performance improvement (from 6.5MB/s to ~24-30MB/s on my IDE test box) when writing out files using O_SYNC. So optimise the batching code by omitting it entirely if the process which is doing a sync write is the same as the one which did the most recent sync write. If that's true, we're unlikely to get any other processes joining the transaction. (Has been in -mm for ages - it took me a long time to get on to performance testing it) Numbers, on write-cache-disabled IDE: /usr/bin/time -p synctest -n 10 -uf -t 1 -p 1 dir-name Unpatched: 40 seconds Patched: 35 seconds Batching disabled: 35 seconds This is the problematic single-process-doing-fsync case. With multiple fsyncing processes the numbers are AFACIT unaltered by the patch. Aside: performance testing and instrumentation shows that the transaction batching almost doesn't help (testing with synctest -n 1 -uf -t 100 -p 10 dir-name on non-writeback-caching IDE). This is because by the time one process is running a synchronous commit, a bunch of other processes already have a transaction handle open, so they're all going to batch into the same transaction anyway. The batching seems to offer maybe 5-10% speedup with this workload, but I'm pretty sure it was more important than that when it was first developed 4-odd years ago... Cc: "Stephen C. Tweedie" <sct@redhat.com> Cc: Benjamin LaHaise <bcrl@kvack.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-05 07:27:54 +00:00
pid_t j_last_sync_writer;
jbd: improve fsync batching There is a flaw with the way jbd handles fsync batching. If we fsync() a file and we were not the last person to run fsync() on this fs then we automatically sleep for 1 jiffie in order to wait for new writers to join into the transaction before forcing the commit. The problem with this is that with really fast storage (ie a Clariion) the time it takes to commit a transaction to disk is way faster than 1 jiffie in most cases, so sleeping means waiting longer with nothing to do than if we just committed the transaction and kept going. Ric Wheeler noticed this when using fs_mark with more than 1 thread, the throughput would plummet as he added more threads. This patch attempts to fix this problem by recording the average time in nanoseconds that it takes to commit a transaction to disk, and what time we started the transaction. If we run an fsync() and we have been running for less time than it takes to commit the transaction to disk, we sleep for the delta amount of time and then commit to disk. We acheive sub-jiffie sleeping using schedule_hrtimeout. This means that the wait time is auto-tuned to the speed of the underlying disk, instead of having this static timeout. I weighted the average according to somebody's comments (Andreas Dilger I think) in order to help normalize random outliers where we take way longer or way less time to commit than the average. I also have a min() check in there to make sure we don't sleep longer than a jiffie in case our storage is super slow, this was requested by Andrew. I unfortunately do not have access to a Clariion, so I had to use a ramdisk to represent a super fast array. I tested with a SATA drive with barrier=1 to make sure there was no regression with local disks, I tested with a 4 way multipathed Apple Xserve RAID array and of course the ramdisk. I ran the following command fs_mark -d /mnt/ext3-test -s 4096 -n 2000 -D 64 -t $i where $i was 2, 4, 8, 16 and 32. I mkfs'ed the fs each time. Here are my results type threads with patch without patch sata 2 24.6 26.3 sata 4 49.2 48.1 sata 8 70.1 67.0 sata 16 104.0 94.1 sata 32 153.6 142.7 xserve 2 246.4 222.0 xserve 4 480.0 440.8 xserve 8 829.5 730.8 xserve 16 1172.7 1026.9 xserve 32 1816.3 1650.5 ramdisk 2 2538.3 1745.6 ramdisk 4 2942.3 661.9 ramdisk 8 2882.5 999.8 ramdisk 16 2738.7 1801.9 ramdisk 32 2541.9 2394.0 Signed-off-by: Josef Bacik <jbacik@redhat.com> Cc: Andreas Dilger <adilger@sun.com> Cc: Arjan van de Ven <arjan@infradead.org> Cc: Ric Wheeler <rwheeler@redhat.com> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-08 02:07:24 +00:00
/*
* the average amount of time in nanoseconds it takes to commit a
* transaction to the disk. [j_state_lock]
*/
u64 j_average_commit_time;
/*
* An opaque pointer to fs-private information. ext3 puts its
* superblock pointer here
*/
void *j_private;
};
/*
* Journal flag definitions
*/
#define JFS_UNMOUNT 0x001 /* Journal thread is being destroyed */
#define JFS_ABORT 0x002 /* Journaling has been aborted for errors. */
#define JFS_ACK_ERR 0x004 /* The errno in the sb has been acked */
#define JFS_FLUSHED 0x008 /* The journal superblock has been flushed */
#define JFS_LOADED 0x010 /* The journal superblock has been loaded */
#define JFS_BARRIER 0x020 /* Use IDE barriers */
#define JFS_ABORT_ON_SYNCDATA_ERR 0x040 /* Abort the journal on file
* data write error in ordered
* mode */
/*
* Function declarations for the journaling transaction and buffer
* management
*/
/* Filing buffers */
extern void journal_unfile_buffer(journal_t *, struct journal_head *);
extern void __journal_unfile_buffer(struct journal_head *);
extern void __journal_refile_buffer(struct journal_head *);
extern void journal_refile_buffer(journal_t *, struct journal_head *);
extern void __journal_file_buffer(struct journal_head *, transaction_t *, int);
extern void __journal_free_buffer(struct journal_head *bh);
extern void journal_file_buffer(struct journal_head *, transaction_t *, int);
extern void __journal_clean_data_list(transaction_t *transaction);
/* Log buffer allocation */
extern struct journal_head * journal_get_descriptor_buffer(journal_t *);
int journal_next_log_block(journal_t *, unsigned int *);
/* Commit management */
extern void journal_commit_transaction(journal_t *);
/* Checkpoint list management */
int __journal_clean_checkpoint_list(journal_t *journal);
int __journal_remove_checkpoint(struct journal_head *);
void __journal_insert_checkpoint(struct journal_head *, transaction_t *);
/* Buffer IO */
extern int
journal_write_metadata_buffer(transaction_t *transaction,
struct journal_head *jh_in,
struct journal_head **jh_out,
unsigned int blocknr);
/* Transaction locking */
extern void __wait_on_journal (journal_t *);
/*
* Journal locking.
*
* We need to lock the journal during transaction state changes so that nobody
* ever tries to take a handle on the running transaction while we are in the
* middle of moving it to the commit phase. j_state_lock does this.
*
* Note that the locking is completely interrupt unsafe. We never touch
* journal structures from interrupts.
*/
static inline handle_t *journal_current_handle(void)
{
return current->journal_info;
}
/* The journaling code user interface:
*
* Create and destroy handles
* Register buffer modifications against the current transaction.
*/
extern handle_t *journal_start(journal_t *, int nblocks);
extern int journal_restart (handle_t *, int nblocks);
extern int journal_extend (handle_t *, int nblocks);
extern int journal_get_write_access(handle_t *, struct buffer_head *);
extern int journal_get_create_access (handle_t *, struct buffer_head *);
extern int journal_get_undo_access(handle_t *, struct buffer_head *);
extern int journal_dirty_data (handle_t *, struct buffer_head *);
extern int journal_dirty_metadata (handle_t *, struct buffer_head *);
extern void journal_release_buffer (handle_t *, struct buffer_head *);
extern int journal_forget (handle_t *, struct buffer_head *);
extern void journal_sync_buffer (struct buffer_head *);
extern void journal_invalidatepage(journal_t *,
struct page *, unsigned long);
extern int journal_try_to_free_buffers(journal_t *, struct page *, gfp_t);
extern int journal_stop(handle_t *);
extern int journal_flush (journal_t *);
extern void journal_lock_updates (journal_t *);
extern void journal_unlock_updates (journal_t *);
extern journal_t * journal_init_dev(struct block_device *bdev,
struct block_device *fs_dev,
int start, int len, int bsize);
extern journal_t * journal_init_inode (struct inode *);
extern int journal_update_format (journal_t *);
extern int journal_check_used_features
(journal_t *, unsigned long, unsigned long, unsigned long);
extern int journal_check_available_features
(journal_t *, unsigned long, unsigned long, unsigned long);
extern int journal_set_features
(journal_t *, unsigned long, unsigned long, unsigned long);
extern int journal_create (journal_t *);
extern int journal_load (journal_t *journal);
jbd: fix error handling for checkpoint io When a checkpointing IO fails, current JBD code doesn't check the error and continue journaling. This means latest metadata can be lost from both the journal and filesystem. This patch leaves the failed metadata blocks in the journal space and aborts journaling in the case of log_do_checkpoint(). To achieve this, we need to do: 1. don't remove the failed buffer from the checkpoint list where in the case of __try_to_free_cp_buf() because it may be released or overwritten by a later transaction 2. log_do_checkpoint() is the last chance, remove the failed buffer from the checkpoint list and abort the journal 3. when checkpointing fails, don't update the journal super block to prevent the journaled contents from being cleaned. For safety, don't update j_tail and j_tail_sequence either 4. when checkpointing fails, notify this error to the ext3 layer so that ext3 don't clear the needs_recovery flag, otherwise the journaled contents are ignored and cleaned in the recovery phase 5. if the recovery fails, keep the needs_recovery flag 6. prevent cleanup_journal_tail() from being called between __journal_drop_transaction() and journal_abort() (a race issue between journal_flush() and __log_wait_for_space() Signed-off-by: Hidehiro Kawai <hidehiro.kawai.ez@hitachi.com> Acked-by: Jan Kara <jack@suse.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-10-22 21:15:00 +00:00
extern int journal_destroy (journal_t *);
extern int journal_recover (journal_t *journal);
extern int journal_wipe (journal_t *, int);
extern int journal_skip_recovery (journal_t *);
extern void journal_update_superblock (journal_t *, int);
extern void journal_abort (journal_t *, int);
extern int journal_errno (journal_t *);
extern void journal_ack_err (journal_t *);
extern int journal_clear_err (journal_t *);
extern int journal_bmap(journal_t *, unsigned int, unsigned int *);
extern int journal_force_commit(journal_t *);
/*
* journal_head management
*/
struct journal_head *journal_add_journal_head(struct buffer_head *bh);
struct journal_head *journal_grab_journal_head(struct buffer_head *bh);
void journal_remove_journal_head(struct buffer_head *bh);
void journal_put_journal_head(struct journal_head *jh);
/*
* handle management
*/
extern struct kmem_cache *jbd_handle_cache;
static inline handle_t *jbd_alloc_handle(gfp_t gfp_flags)
{
return kmem_cache_alloc(jbd_handle_cache, gfp_flags);
}
static inline void jbd_free_handle(handle_t *handle)
{
kmem_cache_free(jbd_handle_cache, handle);
}
/* Primary revoke support */
#define JOURNAL_REVOKE_DEFAULT_HASH 256
extern int journal_init_revoke(journal_t *, int);
extern void journal_destroy_revoke_caches(void);
extern int journal_init_revoke_caches(void);
extern void journal_destroy_revoke(journal_t *);
extern int journal_revoke (handle_t *,
unsigned int, struct buffer_head *);
extern int journal_cancel_revoke(handle_t *, struct journal_head *);
extern void journal_write_revoke_records(journal_t *,
transaction_t *, int);
/* Recovery revoke support */
extern int journal_set_revoke(journal_t *, unsigned int, tid_t);
extern int journal_test_revoke(journal_t *, unsigned int, tid_t);
extern void journal_clear_revoke(journal_t *);
extern void journal_switch_revoke_table(journal_t *journal);
/*
* The log thread user interface:
*
* Request space in the current transaction, and force transaction commit
* transitions on demand.
*/
int __log_space_left(journal_t *); /* Called with journal locked */
int log_start_commit(journal_t *journal, tid_t tid);
int __log_start_commit(journal_t *journal, tid_t tid);
int journal_start_commit(journal_t *journal, tid_t *tid);
int journal_force_commit_nested(journal_t *journal);
int log_wait_commit(journal_t *journal, tid_t tid);
int log_do_checkpoint(journal_t *journal);
int journal_trans_will_send_data_barrier(journal_t *journal, tid_t tid);
void __log_wait_for_space(journal_t *journal);
extern void __journal_drop_transaction(journal_t *, transaction_t *);
extern int cleanup_journal_tail(journal_t *);
/* Debugging code only: */
#define jbd_ENOSYS() \
do { \
printk (KERN_ERR "JBD unimplemented function %s\n", __func__); \
current->state = TASK_UNINTERRUPTIBLE; \
schedule(); \
} while (1)
/*
* is_journal_abort
*
* Simple test wrapper function to test the JFS_ABORT state flag. This
* bit, when set, indicates that we have had a fatal error somewhere,
* either inside the journaling layer or indicated to us by the client
* (eg. ext3), and that we and should not commit any further
* transactions.
*/
static inline int is_journal_aborted(journal_t *journal)
{
return journal->j_flags & JFS_ABORT;
}
static inline int is_handle_aborted(handle_t *handle)
{
if (handle->h_aborted)
return 1;
return is_journal_aborted(handle->h_transaction->t_journal);
}
static inline void journal_abort_handle(handle_t *handle)
{
handle->h_aborted = 1;
}
#endif /* __KERNEL__ */
/* Comparison functions for transaction IDs: perform comparisons using
* modulo arithmetic so that they work over sequence number wraps. */
static inline int tid_gt(tid_t x, tid_t y)
{
int difference = (x - y);
return (difference > 0);
}
static inline int tid_geq(tid_t x, tid_t y)
{
int difference = (x - y);
return (difference >= 0);
}
extern int journal_blocks_per_page(struct inode *inode);
/*
* Return the minimum number of blocks which must be free in the journal
* before a new transaction may be started. Must be called under j_state_lock.
*/
static inline int jbd_space_needed(journal_t *journal)
{
int nblocks = journal->j_max_transaction_buffers;
if (journal->j_committing_transaction)
nblocks += journal->j_committing_transaction->
t_outstanding_credits;
return nblocks;
}
/*
* Definitions which augment the buffer_head layer
*/
/* journaling buffer types */
#define BJ_None 0 /* Not journaled */
#define BJ_SyncData 1 /* Normal data: flush before commit */
#define BJ_Metadata 2 /* Normal journaled metadata */
#define BJ_Forget 3 /* Buffer superseded by this transaction */
#define BJ_IO 4 /* Buffer is for temporary IO use */
#define BJ_Shadow 5 /* Buffer contents being shadowed to the log */
#define BJ_LogCtl 6 /* Buffer contains log descriptors */
#define BJ_Reserved 7 /* Buffer is reserved for access by journal */
#define BJ_Locked 8 /* Locked for I/O during commit */
#define BJ_Types 9
extern int jbd_blocks_per_page(struct inode *inode);
#ifdef __KERNEL__
#define buffer_trace_init(bh) do {} while (0)
#define print_buffer_fields(bh) do {} while (0)
#define print_buffer_trace(bh) do {} while (0)
#define BUFFER_TRACE(bh, info) do {} while (0)
#define BUFFER_TRACE2(bh, bh2, info) do {} while (0)
#define JBUFFER_TRACE(jh, info) do {} while (0)
#endif /* __KERNEL__ */
#endif /* _LINUX_JBD_H */