| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <1157031466.3384.803.camel@quoit.chygwyn.com>
Date: Thu, 31 Aug 2006 14:37:46 +0100
From: Steven Whitehouse <swhiteho@...hat.com>
To: linux-kernel@...r.kernel.org
Cc: Russell Cattelan <cattelan@...hat.com>,
David Teigland <teigland@...hat.com>,
Ingo Molnar <mingo@...e.hu>, hch@...radead.org
Subject: [PATCH 10/16] GFS2: Logging and recovery
[PATCH 10/16] GFS2: Logging and recovery
The journaling and recovery code for GFS2
Signed-off-by: Steven Whitehouse <swhiteho@...hat.com>
Signed-off-by: David Teigland <teigland@...hat.com>
fs/gfs2/log.c | 603 +++++++++++++++++++++++++++++++++++++++++++++++++++++
fs/gfs2/log.h | 61 +++++
fs/gfs2/recovery.c | 573 ++++++++++++++++++++++++++++++++++++++++++++++++++
fs/gfs2/recovery.h | 32 ++
4 files changed, 1269 insertions(+)
--- /dev/null
+++ b/fs/gfs2/log.c
@@ -0,0 +1,603 @@
+/*
+ * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
+ * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved.
+ *
+ * This copyrighted material is made available to anyone wishing to use,
+ * modify, copy, or redistribute it subject to the terms and conditions
+ * of the GNU General Public License v.2.
+ */
+
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/completion.h>
+#include <linux/buffer_head.h>
+#include <linux/gfs2_ondisk.h>
+#include <linux/crc32.h>
+
+#include "gfs2.h"
+#include "lm_interface.h"
+#include "incore.h"
+#include "bmap.h"
+#include "glock.h"
+#include "log.h"
+#include "lops.h"
+#include "meta_io.h"
+#include "util.h"
+#include "dir.h"
+
+#define PULL 1
+
+/**
+ * gfs2_struct2blk - compute stuff
+ * @sdp: the filesystem
+ * @nstruct: the number of structures
+ * @ssize: the size of the structures
+ *
+ * Compute the number of log descriptor blocks needed to hold a certain number
+ * of structures of a certain size.
+ *
+ * Returns: the number of blocks needed (minimum is always 1)
+ */
+
+unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct,
+ unsigned int ssize)
+{
+ unsigned int blks;
+ unsigned int first, second;
+
+ blks = 1;
+ first = (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_log_descriptor)) /
+ ssize;
+
+ if (nstruct > first) {
+ second = (sdp->sd_sb.sb_bsize -
+ sizeof(struct gfs2_meta_header)) / ssize;
+ blks += DIV_ROUND_UP(nstruct - first, second);
+ }
+
+ return blks;
+}
+
+void gfs2_ail1_start(struct gfs2_sbd *sdp, int flags)
+{
+ struct list_head *head = &sdp->sd_ail1_list;
+ uint64_t sync_gen;
+ struct list_head *first, *tmp;
+ struct gfs2_ail *first_ai, *ai;
+
+ gfs2_log_lock(sdp);
+ if (list_empty(head)) {
+ gfs2_log_unlock(sdp);
+ return;
+ }
+ sync_gen = sdp->sd_ail_sync_gen++;
+
+ first = head->prev;
+ first_ai = list_entry(first, struct gfs2_ail, ai_list);
+ first_ai->ai_sync_gen = sync_gen;
+ gfs2_ail1_start_one(sdp, first_ai);
+
+ if (flags & DIO_ALL)
+ first = NULL;
+
+ for (;;) {
+ if (first && (head->prev != first ||
+ gfs2_ail1_empty_one(sdp, first_ai, 0)))
+ break;
+
+ for (tmp = head->prev; tmp != head; tmp = tmp->prev) {
+ ai = list_entry(tmp, struct gfs2_ail, ai_list);
+ if (ai->ai_sync_gen >= sync_gen)
+ continue;
+ ai->ai_sync_gen = sync_gen;
+ gfs2_ail1_start_one(sdp, ai);
+ break;
+ }
+
+ if (tmp == head)
+ break;
+ }
+
+ gfs2_log_unlock(sdp);
+}
+
+int gfs2_ail1_empty(struct gfs2_sbd *sdp, int flags)
+{
+ struct gfs2_ail *ai, *s;
+ int ret;
+
+ gfs2_log_lock(sdp);
+
+ list_for_each_entry_safe_reverse(ai, s, &sdp->sd_ail1_list, ai_list) {
+ if (gfs2_ail1_empty_one(sdp, ai, flags))
+ list_move(&ai->ai_list, &sdp->sd_ail2_list);
+ else if (!(flags & DIO_ALL))
+ break;
+ }
+
+ ret = list_empty(&sdp->sd_ail1_list);
+
+ gfs2_log_unlock(sdp);
+
+ return ret;
+}
+
+static void ail2_empty(struct gfs2_sbd *sdp, unsigned int new_tail)
+{
+ struct gfs2_ail *ai, *safe;
+ unsigned int old_tail = sdp->sd_log_tail;
+ int wrap = (new_tail < old_tail);
+ int a, b, rm;
+
+ gfs2_log_lock(sdp);
+
+ list_for_each_entry_safe(ai, safe, &sdp->sd_ail2_list, ai_list) {
+ a = (old_tail <= ai->ai_first);
+ b = (ai->ai_first < new_tail);
+ rm = (wrap) ? (a || b) : (a && b);
+ if (!rm)
+ continue;
+
+ gfs2_ail2_empty_one(sdp, ai);
+ list_del(&ai->ai_list);
+ gfs2_assert_warn(sdp, list_empty(&ai->ai_ail1_list));
+ gfs2_assert_warn(sdp, list_empty(&ai->ai_ail2_list));
+ kfree(ai);
+ }
+
+ gfs2_log_unlock(sdp);
+}
+
+/**
+ * gfs2_log_reserve - Make a log reservation
+ * @sdp: The GFS2 superblock
+ * @blks: The number of blocks to reserve
+ *
+ * Returns: errno
+ */
+
+int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks)
+{
+ unsigned int try = 0;
+
+ if (gfs2_assert_warn(sdp, blks) ||
+ gfs2_assert_warn(sdp, blks <= sdp->sd_jdesc->jd_blocks))
+ return -EINVAL;
+
+ mutex_lock(&sdp->sd_log_reserve_mutex);
+ gfs2_log_lock(sdp);
+ while(sdp->sd_log_blks_free <= blks) {
+ gfs2_log_unlock(sdp);
+ gfs2_ail1_empty(sdp, 0);
+ gfs2_log_flush(sdp, NULL);
+
+ if (try++)
+ gfs2_ail1_start(sdp, 0);
+ gfs2_log_lock(sdp);
+ }
+ sdp->sd_log_blks_free -= blks;
+ /* printk(KERN_INFO "reserved %u blocks (%u left)\n", blks, sdp->sd_log_blks_free); */
+ gfs2_log_unlock(sdp);
+ mutex_unlock(&sdp->sd_log_reserve_mutex);
+
+ down_read(&sdp->sd_log_flush_lock);
+
+ return 0;
+}
+
+/**
+ * gfs2_log_release - Release a given number of log blocks
+ * @sdp: The GFS2 superblock
+ * @blks: The number of blocks
+ *
+ */
+
+void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks)
+{
+
+ gfs2_log_lock(sdp);
+ sdp->sd_log_blks_free += blks;
+ /* printk(KERN_INFO "released %u blocks (%u left)\n", blks, sdp->sd_log_blks_free); */
+ gfs2_assert_withdraw(sdp,
+ sdp->sd_log_blks_free <= sdp->sd_jdesc->jd_blocks);
+ gfs2_log_unlock(sdp);
+ up_read(&sdp->sd_log_flush_lock);
+}
+
+static uint64_t log_bmap(struct gfs2_sbd *sdp, unsigned int lbn)
+{
+ int new = 0;
+ uint64_t dbn;
+ int error;
+ int bdy;
+
+ error = gfs2_block_map(sdp->sd_jdesc->jd_inode, lbn, &new, &dbn, &bdy);
+ if (!(!error && dbn)) {
+ printk(KERN_INFO "error=%d, dbn=%llu lbn=%u", error, (unsigned long long)dbn, lbn);
+ }
+ gfs2_assert_withdraw(sdp, !error && dbn);
+
+ return dbn;
+}
+
+/**
+ * log_distance - Compute distance between two journal blocks
+ * @sdp: The GFS2 superblock
+ * @newer: The most recent journal block of the pair
+ * @older: The older journal block of the pair
+ *
+ * Compute the distance (in the journal direction) between two
+ * blocks in the journal
+ *
+ * Returns: the distance in blocks
+ */
+
+static inline unsigned int log_distance(struct gfs2_sbd *sdp,
+ unsigned int newer,
+ unsigned int older)
+{
+ int dist;
+
+ dist = newer - older;
+ if (dist < 0)
+ dist += sdp->sd_jdesc->jd_blocks;
+
+ return dist;
+}
+
+static unsigned int current_tail(struct gfs2_sbd *sdp)
+{
+ struct gfs2_ail *ai;
+ unsigned int tail;
+
+ gfs2_log_lock(sdp);
+
+ if (list_empty(&sdp->sd_ail1_list))
+ tail = sdp->sd_log_head;
+ else {
+ ai = list_entry(sdp->sd_ail1_list.prev, struct gfs2_ail,
+ ai_list);
+ tail = ai->ai_first;
+ }
+
+ gfs2_log_unlock(sdp);
+
+ return tail;
+}
+
+static inline void log_incr_head(struct gfs2_sbd *sdp)
+{
+ if (sdp->sd_log_flush_head == sdp->sd_log_tail)
+ gfs2_assert_withdraw(sdp,
+ sdp->sd_log_flush_head == sdp->sd_log_head);
+
+ if (++sdp->sd_log_flush_head == sdp->sd_jdesc->jd_blocks) {
+ sdp->sd_log_flush_head = 0;
+ sdp->sd_log_flush_wrapped = 1;
+ }
+}
+
+/**
+ * gfs2_log_get_buf - Get and initialize a buffer to use for log control data
+ * @sdp: The GFS2 superblock
+ *
+ * Returns: the buffer_head
+ */
+
+struct buffer_head *gfs2_log_get_buf(struct gfs2_sbd *sdp)
+{
+ uint64_t blkno = log_bmap(sdp, sdp->sd_log_flush_head);
+ struct gfs2_log_buf *lb;
+ struct buffer_head *bh;
+
+ lb = kzalloc(sizeof(struct gfs2_log_buf), GFP_NOFS | __GFP_NOFAIL);
+ list_add(&lb->lb_list, &sdp->sd_log_flush_list);
+
+ bh = lb->lb_bh = sb_getblk(sdp->sd_vfs, blkno);
+ lock_buffer(bh);
+ memset(bh->b_data, 0, bh->b_size);
+ set_buffer_uptodate(bh);
+ clear_buffer_dirty(bh);
+ unlock_buffer(bh);
+
+ log_incr_head(sdp);
+
+ return bh;
+}
+
+/**
+ * gfs2_log_fake_buf - Build a fake buffer head to write metadata buffer to log
+ * @sdp: the filesystem
+ * @data: the data the buffer_head should point to
+ *
+ * Returns: the log buffer descriptor
+ */
+
+struct buffer_head *gfs2_log_fake_buf(struct gfs2_sbd *sdp,
+ struct buffer_head *real)
+{
+ uint64_t blkno = log_bmap(sdp, sdp->sd_log_flush_head);
+ struct gfs2_log_buf *lb;
+ struct buffer_head *bh;
+
+ lb = kzalloc(sizeof(struct gfs2_log_buf), GFP_NOFS | __GFP_NOFAIL);
+ list_add(&lb->lb_list, &sdp->sd_log_flush_list);
+ lb->lb_real = real;
+
+ bh = lb->lb_bh = alloc_buffer_head(GFP_NOFS | __GFP_NOFAIL);
+ atomic_set(&bh->b_count, 1);
+ bh->b_state = (1 << BH_Mapped) | (1 << BH_Uptodate);
+ set_bh_page(bh, real->b_page, bh_offset(real));
+ bh->b_blocknr = blkno;
+ bh->b_size = sdp->sd_sb.sb_bsize;
+ bh->b_bdev = sdp->sd_vfs->s_bdev;
+
+ log_incr_head(sdp);
+
+ return bh;
+}
+
+static void log_pull_tail(struct gfs2_sbd *sdp, unsigned int new_tail, int pull)
+{
+ unsigned int dist = log_distance(sdp, new_tail, sdp->sd_log_tail);
+
+ ail2_empty(sdp, new_tail);
+
+ gfs2_log_lock(sdp);
+ sdp->sd_log_blks_free += dist - ((pull) ? 1 : 0);
+ /* printk(KERN_INFO "pull tail refunding %u blocks (%u left) pull=%d\n", dist - ((pull) ? 1 : 0), sdp->sd_log_blks_free, pull); */
+ gfs2_assert_withdraw(sdp,
+ sdp->sd_log_blks_free <= sdp->sd_jdesc->jd_blocks);
+ gfs2_log_unlock(sdp);
+
+ sdp->sd_log_tail = new_tail;
+}
+
+/**
+ * log_write_header - Get and initialize a journal header buffer
+ * @sdp: The GFS2 superblock
+ *
+ * Returns: the initialized log buffer descriptor
+ */
+
+static void log_write_header(struct gfs2_sbd *sdp, uint32_t flags, int pull)
+{
+ uint64_t blkno = log_bmap(sdp, sdp->sd_log_flush_head);
+ struct buffer_head *bh;
+ struct gfs2_log_header *lh;
+ unsigned int tail;
+ uint32_t hash;
+
+ /* printk(KERN_INFO "log write header start (flags=%08x, pull=%d)\n", flags, pull); */
+
+ bh = sb_getblk(sdp->sd_vfs, blkno);
+ lock_buffer(bh);
+ memset(bh->b_data, 0, bh->b_size);
+ set_buffer_uptodate(bh);
+ clear_buffer_dirty(bh);
+ unlock_buffer(bh);
+
+ gfs2_ail1_empty(sdp, 0);
+ tail = current_tail(sdp);
+
+ lh = (struct gfs2_log_header *)bh->b_data;
+ memset(lh, 0, sizeof(struct gfs2_log_header));
+ lh->lh_header.mh_magic = cpu_to_be32(GFS2_MAGIC);
+ lh->lh_header.mh_type = cpu_to_be32(GFS2_METATYPE_LH);
+ lh->lh_header.mh_format = cpu_to_be32(GFS2_FORMAT_LH);
+ lh->lh_sequence = cpu_to_be64(sdp->sd_log_sequence++);
+ lh->lh_flags = cpu_to_be32(flags);
+ lh->lh_tail = cpu_to_be32(tail);
+ lh->lh_blkno = cpu_to_be32(sdp->sd_log_flush_head);
+ hash = gfs2_disk_hash(bh->b_data, sizeof(struct gfs2_log_header));
+ lh->lh_hash = cpu_to_be32(hash);
+
+ set_buffer_dirty(bh);
+ if (sync_dirty_buffer(bh))
+ gfs2_io_error_bh(sdp, bh);
+ brelse(bh);
+
+ if (sdp->sd_log_tail != tail)
+ log_pull_tail(sdp, tail, pull);
+ else
+ gfs2_assert_withdraw(sdp, !pull);
+
+ sdp->sd_log_idle = (tail == sdp->sd_log_flush_head);
+ log_incr_head(sdp);
+
+ /* printk(KERN_INFO "log write header out\n"); */
+}
+
+static void log_flush_commit(struct gfs2_sbd *sdp)
+{
+ struct list_head *head = &sdp->sd_log_flush_list;
+ struct gfs2_log_buf *lb;
+ struct buffer_head *bh;
+#if 0
+ unsigned int d;
+
+ d = log_distance(sdp, sdp->sd_log_flush_head, sdp->sd_log_head);
+
+ gfs2_assert_withdraw(sdp, d + 1 == sdp->sd_log_blks_reserved);
+#endif
+
+ while (!list_empty(head)) {
+ lb = list_entry(head->next, struct gfs2_log_buf, lb_list);
+ list_del(&lb->lb_list);
+ bh = lb->lb_bh;
+
+ wait_on_buffer(bh);
+ if (!buffer_uptodate(bh))
+ gfs2_io_error_bh(sdp, bh);
+ if (lb->lb_real) {
+ while (atomic_read(&bh->b_count) != 1) /* Grrrr... */
+ schedule();
+ free_buffer_head(bh);
+ } else
+ brelse(bh);
+ kfree(lb);
+ }
+
+ log_write_header(sdp, 0, 0);
+}
+
+/**
+ * gfs2_log_flush - flush incore transaction(s)
+ * @sdp: the filesystem
+ * @gl: The glock structure to flush. If NULL, flush the whole incore log
+ *
+ */
+
+void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl)
+{
+ struct gfs2_ail *ai;
+
+ down_write(&sdp->sd_log_flush_lock);
+
+ if (gl) {
+ gfs2_log_lock(sdp);
+ if (list_empty(&gl->gl_le.le_list)) {
+ gfs2_log_unlock(sdp);
+ up_write(&sdp->sd_log_flush_lock);
+ return;
+ }
+ gfs2_log_unlock(sdp);
+ }
+
+ ai = kzalloc(sizeof(struct gfs2_ail), GFP_NOFS | __GFP_NOFAIL);
+ INIT_LIST_HEAD(&ai->ai_ail1_list);
+ INIT_LIST_HEAD(&ai->ai_ail2_list);
+
+ gfs2_assert_withdraw(sdp,
+ sdp->sd_log_num_buf == sdp->sd_log_commited_buf);
+ gfs2_assert_withdraw(sdp,
+ sdp->sd_log_num_revoke == sdp->sd_log_commited_revoke);
+
+ sdp->sd_log_flush_head = sdp->sd_log_head;
+ sdp->sd_log_flush_wrapped = 0;
+ ai->ai_first = sdp->sd_log_flush_head;
+
+ lops_before_commit(sdp);
+ if (!list_empty(&sdp->sd_log_flush_list))
+ log_flush_commit(sdp);
+ else if (sdp->sd_log_tail != current_tail(sdp) && !sdp->sd_log_idle)
+ log_write_header(sdp, 0, PULL);
+ lops_after_commit(sdp, ai);
+ sdp->sd_log_head = sdp->sd_log_flush_head;
+
+ /* printk(KERN_INFO "sd_log_num_hdrs %u\n", sdp->sd_log_num_hdrs); */
+ sdp->sd_log_blks_free -= sdp->sd_log_num_hdrs;
+
+ sdp->sd_log_blks_reserved =
+ sdp->sd_log_commited_buf =
+ sdp->sd_log_num_hdrs =
+ sdp->sd_log_commited_revoke = 0;
+
+ gfs2_log_lock(sdp);
+ if (!list_empty(&ai->ai_ail1_list)) {
+ list_add(&ai->ai_list, &sdp->sd_ail1_list);
+ ai = NULL;
+ }
+ gfs2_log_unlock(sdp);
+
+ sdp->sd_vfs->s_dirt = 0;
+ up_write(&sdp->sd_log_flush_lock);
+
+ kfree(ai);
+}
+
+static void log_refund(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
+{
+ unsigned int reserved = 0;
+ unsigned int old;
+
+ gfs2_log_lock(sdp);
+
+ sdp->sd_log_commited_buf += tr->tr_num_buf_new - tr->tr_num_buf_rm;
+ gfs2_assert_withdraw(sdp, ((int)sdp->sd_log_commited_buf) >= 0);
+ sdp->sd_log_commited_revoke += tr->tr_num_revoke - tr->tr_num_revoke_rm;
+ gfs2_assert_withdraw(sdp, ((int)sdp->sd_log_commited_revoke) >= 0);
+
+ if (sdp->sd_log_commited_buf)
+ reserved += sdp->sd_log_commited_buf;
+ if (sdp->sd_log_commited_revoke)
+ reserved += gfs2_struct2blk(sdp, sdp->sd_log_commited_revoke,
+ sizeof(uint64_t));
+ if (reserved)
+ reserved++;
+
+ old = sdp->sd_log_blks_free;
+ sdp->sd_log_blks_free += tr->tr_reserved -
+ (reserved - sdp->sd_log_blks_reserved);
+
+ gfs2_assert_withdraw(sdp, sdp->sd_log_blks_free >= old);
+ gfs2_assert_withdraw(sdp,
+ sdp->sd_log_blks_free <= sdp->sd_jdesc->jd_blocks +
+ sdp->sd_log_num_hdrs);
+
+ sdp->sd_log_blks_reserved = reserved;
+
+ gfs2_log_unlock(sdp);
+}
+
+/**
+ * gfs2_log_commit - Commit a transaction to the log
+ * @sdp: the filesystem
+ * @tr: the transaction
+ *
+ * Returns: errno
+ */
+
+void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
+{
+ log_refund(sdp, tr);
+ lops_incore_commit(sdp, tr);
+
+ sdp->sd_vfs->s_dirt = 1;
+ up_read(&sdp->sd_log_flush_lock);
+
+ gfs2_log_lock(sdp);
+ if (sdp->sd_log_num_buf > gfs2_tune_get(sdp, gt_incore_log_blocks)) {
+ gfs2_log_unlock(sdp);
+ gfs2_log_flush(sdp, NULL);
+ } else
+ gfs2_log_unlock(sdp);
+}
+
+/**
+ * gfs2_log_shutdown - write a shutdown header into a journal
+ * @sdp: the filesystem
+ *
+ */
+
+void gfs2_log_shutdown(struct gfs2_sbd *sdp)
+{
+ down_write(&sdp->sd_log_flush_lock);
+
+ gfs2_assert_withdraw(sdp, !sdp->sd_log_blks_reserved);
+ gfs2_assert_withdraw(sdp, !sdp->sd_log_num_gl);
+ gfs2_assert_withdraw(sdp, !sdp->sd_log_num_buf);
+ gfs2_assert_withdraw(sdp, !sdp->sd_log_num_jdata);
+ gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke);
+ gfs2_assert_withdraw(sdp, !sdp->sd_log_num_rg);
+ gfs2_assert_withdraw(sdp, !sdp->sd_log_num_databuf);
+ gfs2_assert_withdraw(sdp, !sdp->sd_log_num_hdrs);
+ gfs2_assert_withdraw(sdp, list_empty(&sdp->sd_ail1_list));
+
+ sdp->sd_log_flush_head = sdp->sd_log_head;
+ sdp->sd_log_flush_wrapped = 0;
+
+ log_write_header(sdp, GFS2_LOG_HEAD_UNMOUNT, 0);
+
+ /* printk(KERN_INFO "sd_log_blks_free %u, sd_jdesc->jd_blocks %u\n", sdp->sd_log_blks_free, sdp->sd_jdesc->jd_blocks); */
+ gfs2_assert_warn(sdp, sdp->sd_log_blks_free == sdp->sd_jdesc->jd_blocks);
+ gfs2_assert_warn(sdp, sdp->sd_log_head == sdp->sd_log_tail);
+ gfs2_assert_warn(sdp, list_empty(&sdp->sd_ail2_list));
+
+ sdp->sd_log_head = sdp->sd_log_flush_head;
+ sdp->sd_log_tail = sdp->sd_log_head;
+
+ up_write(&sdp->sd_log_flush_lock);
+}
+
--- /dev/null
+++ b/fs/gfs2/log.h
@@ -0,0 +1,61 @@
+/*
+ * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
+ * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved.
+ *
+ * This copyrighted material is made available to anyone wishing to use,
+ * modify, copy, or redistribute it subject to the terms and conditions
+ * of the GNU General Public License v.2.
+ */
+
+#ifndef __LOG_DOT_H__
+#define __LOG_DOT_H__
+
+/**
+ * gfs2_log_lock - acquire the right to mess with the log manager
+ * @sdp: the filesystem
+ *
+ */
+
+static inline void gfs2_log_lock(struct gfs2_sbd *sdp)
+{
+ spin_lock(&sdp->sd_log_lock);
+}
+
+/**
+ * gfs2_log_unlock - release the right to mess with the log manager
+ * @sdp: the filesystem
+ *
+ */
+
+static inline void gfs2_log_unlock(struct gfs2_sbd *sdp)
+{
+ spin_unlock(&sdp->sd_log_lock);
+}
+
+static inline void gfs2_log_pointers_init(struct gfs2_sbd *sdp,
+ unsigned int value)
+{
+ if (++value == sdp->sd_jdesc->jd_blocks) {
+ value = 0;
+ }
+ sdp->sd_log_head = sdp->sd_log_tail = value;
+}
+
+unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct,
+ unsigned int ssize);
+
+void gfs2_ail1_start(struct gfs2_sbd *sdp, int flags);
+int gfs2_ail1_empty(struct gfs2_sbd *sdp, int flags);
+
+int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks);
+void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks);
+
+struct buffer_head *gfs2_log_get_buf(struct gfs2_sbd *sdp);
+struct buffer_head *gfs2_log_fake_buf(struct gfs2_sbd *sdp,
+ struct buffer_head *real);
+void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl);
+void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *trans);
+
+void gfs2_log_shutdown(struct gfs2_sbd *sdp);
+
+#endif /* __LOG_DOT_H__ */
--- /dev/null
+++ b/fs/gfs2/recovery.c
@@ -0,0 +1,573 @@
+/*
+ * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
+ * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved.
+ *
+ * This copyrighted material is made available to anyone wishing to use,
+ * modify, copy, or redistribute it subject to the terms and conditions
+ * of the GNU General Public License v.2.
+ */
+
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/completion.h>
+#include <linux/buffer_head.h>
+#include <linux/gfs2_ondisk.h>
+#include <linux/crc32.h>
+
+#include "gfs2.h"
+#include "lm_interface.h"
+#include "incore.h"
+#include "bmap.h"
+#include "glock.h"
+#include "glops.h"
+#include "lm.h"
+#include "lops.h"
+#include "meta_io.h"
+#include "recovery.h"
+#include "super.h"
+#include "util.h"
+#include "dir.h"
+
+int gfs2_replay_read_block(struct gfs2_jdesc *jd, unsigned int blk,
+ struct buffer_head **bh)
+{
+ struct gfs2_inode *ip = GFS2_I(jd->jd_inode);
+ struct gfs2_glock *gl = ip->i_gl;
+ int new = 0;
+ uint64_t dblock;
+ uint32_t extlen;
+ int error;
+
+ error = gfs2_extent_map(&ip->i_inode, blk, &new, &dblock, &extlen);
+ if (error)
+ return error;
+ if (!dblock) {
+ gfs2_consist_inode(ip);
+ return -EIO;
+ }
+
+ gfs2_meta_ra(gl, dblock, extlen);
+ error = gfs2_meta_read(gl, dblock, DIO_START | DIO_WAIT, bh);
+
+ return error;
+}
+
+int gfs2_revoke_add(struct gfs2_sbd *sdp, uint64_t blkno, unsigned int where)
+{
+ struct list_head *head = &sdp->sd_revoke_list;
+ struct gfs2_revoke_replay *rr;
+ int found = 0;
+
+ list_for_each_entry(rr, head, rr_list) {
+ if (rr->rr_blkno == blkno) {
+ found = 1;
+ break;
+ }
+ }
+
+ if (found) {
+ rr->rr_where = where;
+ return 0;
+ }
+
+ rr = kmalloc(sizeof(struct gfs2_revoke_replay), GFP_KERNEL);
+ if (!rr)
+ return -ENOMEM;
+
+ rr->rr_blkno = blkno;
+ rr->rr_where = where;
+ list_add(&rr->rr_list, head);
+
+ return 1;
+}
+
+int gfs2_revoke_check(struct gfs2_sbd *sdp, uint64_t blkno, unsigned int where)
+{
+ struct gfs2_revoke_replay *rr;
+ int wrap, a, b, revoke;
+ int found = 0;
+
+ list_for_each_entry(rr, &sdp->sd_revoke_list, rr_list) {
+ if (rr->rr_blkno == blkno) {
+ found = 1;
+ break;
+ }
+ }
+
+ if (!found)
+ return 0;
+
+ wrap = (rr->rr_where < sdp->sd_replay_tail);
+ a = (sdp->sd_replay_tail < where);
+ b = (where < rr->rr_where);
+ revoke = (wrap) ? (a || b) : (a && b);
+
+ return revoke;
+}
+
+void gfs2_revoke_clean(struct gfs2_sbd *sdp)
+{
+ struct list_head *head = &sdp->sd_revoke_list;
+ struct gfs2_revoke_replay *rr;
+
+ while (!list_empty(head)) {
+ rr = list_entry(head->next, struct gfs2_revoke_replay, rr_list);
+ list_del(&rr->rr_list);
+ kfree(rr);
+ }
+}
+
+/**
+ * get_log_header - read the log header for a given segment
+ * @jd: the journal
+ * @blk: the block to look at
+ * @lh: the log header to return
+ *
+ * Read the log header for a given segement in a given journal. Do a few
+ * sanity checks on it.
+ *
+ * Returns: 0 on success,
+ * 1 if the header was invalid or incomplete,
+ * errno on error
+ */
+
+static int get_log_header(struct gfs2_jdesc *jd, unsigned int blk,
+ struct gfs2_log_header *head)
+{
+ struct buffer_head *bh;
+ struct gfs2_log_header lh;
+ uint32_t hash;
+ int error;
+
+ error = gfs2_replay_read_block(jd, blk, &bh);
+ if (error)
+ return error;
+
+ memcpy(&lh, bh->b_data, sizeof(struct gfs2_log_header));
+ lh.lh_hash = 0;
+ hash = gfs2_disk_hash((char *)&lh, sizeof(struct gfs2_log_header));
+ gfs2_log_header_in(&lh, bh->b_data);
+
+ brelse(bh);
+
+ if (lh.lh_header.mh_magic != GFS2_MAGIC ||
+ lh.lh_header.mh_type != GFS2_METATYPE_LH ||
+ lh.lh_blkno != blk || lh.lh_hash != hash)
+ return 1;
+
+ *head = lh;
+
+ return 0;
+}
+
+/**
+ * find_good_lh - find a good log header
+ * @jd: the journal
+ * @blk: the segment to start searching from
+ * @lh: the log header to fill in
+ * @forward: if true search forward in the log, else search backward
+ *
+ * Call get_log_header() to get a log header for a segment, but if the
+ * segment is bad, either scan forward or backward until we find a good one.
+ *
+ * Returns: errno
+ */
+
+static int find_good_lh(struct gfs2_jdesc *jd, unsigned int *blk,
+ struct gfs2_log_header *head)
+{
+ unsigned int orig_blk = *blk;
+ int error;
+
+ for (;;) {
+ error = get_log_header(jd, *blk, head);
+ if (error <= 0)
+ return error;
+
+ if (++*blk == jd->jd_blocks)
+ *blk = 0;
+
+ if (*blk == orig_blk) {
+ gfs2_consist_inode(GFS2_I(jd->jd_inode));
+ return -EIO;
+ }
+ }
+}
+
+/**
+ * jhead_scan - make sure we've found the head of the log
+ * @jd: the journal
+ * @head: this is filled in with the log descriptor of the head
+ *
+ * At this point, seg and lh should be either the head of the log or just
+ * before. Scan forward until we find the head.
+ *
+ * Returns: errno
+ */
+
+static int jhead_scan(struct gfs2_jdesc *jd, struct gfs2_log_header *head)
+{
+ unsigned int blk = head->lh_blkno;
+ struct gfs2_log_header lh;
+ int error;
+
+ for (;;) {
+ if (++blk == jd->jd_blocks)
+ blk = 0;
+
+ error = get_log_header(jd, blk, &lh);
+ if (error < 0)
+ return error;
+ if (error == 1)
+ continue;
+
+ if (lh.lh_sequence == head->lh_sequence) {
+ gfs2_consist_inode(GFS2_I(jd->jd_inode));
+ return -EIO;
+ }
+ if (lh.lh_sequence < head->lh_sequence)
+ break;
+
+ *head = lh;
+ }
+
+ return 0;
+}
+
+/**
+ * gfs2_find_jhead - find the head of a log
+ * @jd: the journal
+ * @head: the log descriptor for the head of the log is returned here
+ *
+ * Do a binary search of a journal and find the valid log entry with the
+ * highest sequence number. (i.e. the log head)
+ *
+ * Returns: errno
+ */
+
+int gfs2_find_jhead(struct gfs2_jdesc *jd, struct gfs2_log_header *head)
+{
+ struct gfs2_log_header lh_1, lh_m;
+ uint32_t blk_1, blk_2, blk_m;
+ int error;
+
+ blk_1 = 0;
+ blk_2 = jd->jd_blocks - 1;
+
+ for (;;) {
+ blk_m = (blk_1 + blk_2) / 2;
+
+ error = find_good_lh(jd, &blk_1, &lh_1);
+ if (error)
+ return error;
+
+ error = find_good_lh(jd, &blk_m, &lh_m);
+ if (error)
+ return error;
+
+ if (blk_1 == blk_m || blk_m == blk_2)
+ break;
+
+ if (lh_1.lh_sequence <= lh_m.lh_sequence)
+ blk_1 = blk_m;
+ else
+ blk_2 = blk_m;
+ }
+
+ error = jhead_scan(jd, &lh_1);
+ if (error)
+ return error;
+
+ *head = lh_1;
+
+ return error;
+}
+
+/**
+ * foreach_descriptor - go through the active part of the log
+ * @jd: the journal
+ * @start: the first log header in the active region
+ * @end: the last log header (don't process the contents of this entry))
+ *
+ * Call a given function once for every log descriptor in the active
+ * portion of the log.
+ *
+ * Returns: errno
+ */
+
+static int foreach_descriptor(struct gfs2_jdesc *jd, unsigned int start,
+ unsigned int end, int pass)
+{
+ struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode);
+ struct buffer_head *bh;
+ struct gfs2_log_descriptor *ld;
+ int error = 0;
+ u32 length;
+ __be64 *ptr;
+ unsigned int offset = sizeof(struct gfs2_log_descriptor);
+ offset += (sizeof(__be64)-1);
+ offset &= ~(sizeof(__be64)-1);
+
+ while (start != end) {
+ error = gfs2_replay_read_block(jd, start, &bh);
+ if (error)
+ return error;
+ if (gfs2_meta_check(sdp, bh)) {
+ brelse(bh);
+ return -EIO;
+ }
+ ld = (struct gfs2_log_descriptor *)bh->b_data;
+ length = be32_to_cpu(ld->ld_length);
+
+ if (be32_to_cpu(ld->ld_header.mh_type) == GFS2_METATYPE_LH) {
+ struct gfs2_log_header lh;
+ error = get_log_header(jd, start, &lh);
+ if (!error) {
+ gfs2_replay_incr_blk(sdp, &start);
+ brelse(bh);
+ continue;
+ }
+ if (error == 1) {
+ gfs2_consist_inode(GFS2_I(jd->jd_inode));
+ error = -EIO;
+ }
+ brelse(bh);
+ return error;
+ } else if (gfs2_metatype_check(sdp, bh, GFS2_METATYPE_LD)) {
+ brelse(bh);
+ return -EIO;
+ }
+ ptr = (__be64 *)(bh->b_data + offset);
+ error = lops_scan_elements(jd, start, ld, ptr, pass);
+ if (error) {
+ brelse(bh);
+ return error;
+ }
+
+ while (length--)
+ gfs2_replay_incr_blk(sdp, &start);
+
+ brelse(bh);
+ }
+
+ return 0;
+}
+
+/**
+ * clean_journal - mark a dirty journal as being clean
+ * @sdp: the filesystem
+ * @jd: the journal
+ * @gl: the journal's glock
+ * @head: the head journal to start from
+ *
+ * Returns: errno
+ */
+
+static int clean_journal(struct gfs2_jdesc *jd, struct gfs2_log_header *head)
+{
+ struct gfs2_inode *ip = GFS2_I(jd->jd_inode);
+ struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode);
+ unsigned int lblock;
+ int new = 0;
+ uint64_t dblock;
+ struct gfs2_log_header *lh;
+ uint32_t hash;
+ struct buffer_head *bh;
+ int error;
+ int boundary;
+
+ lblock = head->lh_blkno;
+ gfs2_replay_incr_blk(sdp, &lblock);
+ error = gfs2_block_map(&ip->i_inode, lblock, &new, &dblock, &boundary);
+ if (error)
+ return error;
+ if (!dblock) {
+ gfs2_consist_inode(ip);
+ return -EIO;
+ }
+
+ bh = sb_getblk(sdp->sd_vfs, dblock);
+ lock_buffer(bh);
+ memset(bh->b_data, 0, bh->b_size);
+ set_buffer_uptodate(bh);
+ clear_buffer_dirty(bh);
+ unlock_buffer(bh);
+
+ lh = (struct gfs2_log_header *)bh->b_data;
+ memset(lh, 0, sizeof(struct gfs2_log_header));
+ lh->lh_header.mh_magic = cpu_to_be32(GFS2_MAGIC);
+ lh->lh_header.mh_type = cpu_to_be32(GFS2_METATYPE_LH);
+ lh->lh_header.mh_format = cpu_to_be32(GFS2_FORMAT_LH);
+ lh->lh_sequence = cpu_to_be64(head->lh_sequence + 1);
+ lh->lh_flags = cpu_to_be32(GFS2_LOG_HEAD_UNMOUNT);
+ lh->lh_blkno = cpu_to_be32(lblock);
+ hash = gfs2_disk_hash((const char *)lh, sizeof(struct gfs2_log_header));
+ lh->lh_hash = cpu_to_be32(hash);
+
+ set_buffer_dirty(bh);
+ if (sync_dirty_buffer(bh))
+ gfs2_io_error_bh(sdp, bh);
+ brelse(bh);
+
+ return error;
+}
+
+/**
+ * gfs2_recover_journal - recovery a given journal
+ * @jd: the struct gfs2_jdesc describing the journal
+ *
+ * Acquire the journal's lock, check to see if the journal is clean, and
+ * do recovery if necessary.
+ *
+ * Returns: errno
+ */
+
+int gfs2_recover_journal(struct gfs2_jdesc *jd)
+{
+ struct gfs2_inode *ip = GFS2_I(jd->jd_inode);
+ struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode);
+ struct gfs2_log_header head;
+ struct gfs2_holder j_gh, ji_gh, t_gh;
+ unsigned long t;
+ int ro = 0;
+ unsigned int pass;
+ int error;
+
+ if (jd->jd_jid != sdp->sd_lockstruct.ls_jid) {
+ fs_info(sdp, "jid=%u: Trying to acquire journal lock...\n",
+ jd->jd_jid);
+
+ /* Aquire the journal lock so we can do recovery */
+
+ error = gfs2_glock_nq_num(sdp, jd->jd_jid, &gfs2_journal_glops,
+ LM_ST_EXCLUSIVE,
+ LM_FLAG_NOEXP | LM_FLAG_TRY | GL_NOCACHE,
+ &j_gh);
+ switch (error) {
+ case 0:
+ break;
+
+ case GLR_TRYFAILED:
+ fs_info(sdp, "jid=%u: Busy\n", jd->jd_jid);
+ error = 0;
+
+ default:
+ goto fail;
+ };
+
+ error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED,
+ LM_FLAG_NOEXP, &ji_gh);
+ if (error)
+ goto fail_gunlock_j;
+ } else {
+ fs_info(sdp, "jid=%u, already locked for use\n", jd->jd_jid);
+ }
+
+ fs_info(sdp, "jid=%u: Looking at journal...\n", jd->jd_jid);
+
+ error = gfs2_jdesc_check(jd);
+ if (error)
+ goto fail_gunlock_ji;
+
+ error = gfs2_find_jhead(jd, &head);
+ if (error)
+ goto fail_gunlock_ji;
+
+ if (!(head.lh_flags & GFS2_LOG_HEAD_UNMOUNT)) {
+ fs_info(sdp, "jid=%u: Acquiring the transaction lock...\n",
+ jd->jd_jid);
+
+ t = jiffies;
+
+ /* Acquire a shared hold on the transaction lock */
+
+ error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_SHARED,
+ LM_FLAG_NOEXP | LM_FLAG_PRIORITY |
+ GL_NOCANCEL | GL_NOCACHE, &t_gh);
+ if (error)
+ goto fail_gunlock_ji;
+
+ if (test_bit(SDF_JOURNAL_CHECKED, &sdp->sd_flags)) {
+ if (!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags))
+ ro = 1;
+ } else {
+ if (sdp->sd_vfs->s_flags & MS_RDONLY)
+ ro = 1;
+ }
+
+ if (ro) {
+ fs_warn(sdp, "jid=%u: Can't replay: read-only FS\n",
+ jd->jd_jid);
+ error = -EROFS;
+ goto fail_gunlock_tr;
+ }
+
+ fs_info(sdp, "jid=%u: Replaying journal...\n", jd->jd_jid);
+
+ for (pass = 0; pass < 2; pass++) {
+ lops_before_scan(jd, &head, pass);
+ error = foreach_descriptor(jd, head.lh_tail,
+ head.lh_blkno, pass);
+ lops_after_scan(jd, error, pass);
+ if (error)
+ goto fail_gunlock_tr;
+ }
+
+ error = clean_journal(jd, &head);
+ if (error)
+ goto fail_gunlock_tr;
+
+ gfs2_glock_dq_uninit(&t_gh);
+ t = DIV_ROUND_UP(jiffies - t, HZ);
+ fs_info(sdp, "jid=%u: Journal replayed in %lus\n",
+ jd->jd_jid, t);
+ }
+
+ if (jd->jd_jid != sdp->sd_lockstruct.ls_jid)
+ gfs2_glock_dq_uninit(&ji_gh);
+
+ gfs2_lm_recovery_done(sdp, jd->jd_jid, LM_RD_SUCCESS);
+
+ if (jd->jd_jid != sdp->sd_lockstruct.ls_jid)
+ gfs2_glock_dq_uninit(&j_gh);
+
+ fs_info(sdp, "jid=%u: Done\n", jd->jd_jid);
+ return 0;
+
+fail_gunlock_tr:
+ gfs2_glock_dq_uninit(&t_gh);
+fail_gunlock_ji:
+ if (jd->jd_jid != sdp->sd_lockstruct.ls_jid) {
+ gfs2_glock_dq_uninit(&ji_gh);
+fail_gunlock_j:
+ gfs2_glock_dq_uninit(&j_gh);
+ }
+
+ fs_info(sdp, "jid=%u: %s\n", jd->jd_jid, (error) ? "Failed" : "Done");
+
+fail:
+ gfs2_lm_recovery_done(sdp, jd->jd_jid, LM_RD_GAVEUP);
+ return error;
+}
+
+/**
+ * gfs2_check_journals - Recover any dirty journals
+ * @sdp: the filesystem
+ *
+ */
+
+void gfs2_check_journals(struct gfs2_sbd *sdp)
+{
+ struct gfs2_jdesc *jd;
+
+ for (;;) {
+ jd = gfs2_jdesc_find_dirty(sdp);
+ if (!jd)
+ break;
+
+ if (jd != sdp->sd_jdesc)
+ gfs2_recover_journal(jd);
+ }
+}
+
--- /dev/null
+++ b/fs/gfs2/recovery.h
@@ -0,0 +1,32 @@
+/*
+ * Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
+ * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved.
+ *
+ * This copyrighted material is made available to anyone wishing to use,
+ * modify, copy, or redistribute it subject to the terms and conditions
+ * of the GNU General Public License v.2.
+ */
+
+#ifndef __RECOVERY_DOT_H__
+#define __RECOVERY_DOT_H__
+
+static inline void gfs2_replay_incr_blk(struct gfs2_sbd *sdp, unsigned int *blk)
+{
+ if (++*blk == sdp->sd_jdesc->jd_blocks)
+ *blk = 0;
+}
+
+int gfs2_replay_read_block(struct gfs2_jdesc *jd, unsigned int blk,
+ struct buffer_head **bh);
+
+int gfs2_revoke_add(struct gfs2_sbd *sdp, uint64_t blkno, unsigned int where);
+int gfs2_revoke_check(struct gfs2_sbd *sdp, uint64_t blkno, unsigned int where);
+void gfs2_revoke_clean(struct gfs2_sbd *sdp);
+
+int gfs2_find_jhead(struct gfs2_jdesc *jd,
+ struct gfs2_log_header *head);
+int gfs2_recover_journal(struct gfs2_jdesc *gfs2_jd);
+void gfs2_check_journals(struct gfs2_sbd *sdp);
+
+#endif /* __RECOVERY_DOT_H__ */
+
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists