X-Git-Url: https://err.no/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=fs%2Fgfs2%2Flog.c;h=291415ddfe51cb291ccd66b23703e63a58b089d7;hb=ecd27b92fbb41f779d857632a69bd45dbaf0f915;hp=0b26d6a74118fbbd818133d5e4bc3e6b0de5aa11;hpb=ed3865079b573ef55dc13ab0bfb242ed5ebab4c1;p=linux-2.6

diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c
index 0b26d6a741..291415ddfe 100644
--- a/fs/gfs2/log.c
+++ b/fs/gfs2/log.c
@@ -1,10 +1,10 @@
 /*
  * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
- * Copyright (C) 2004-2005 Red Hat, Inc.  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.
+ * of the GNU General Public License version 2.
  */
 
 #include <linux/sched.h>
@@ -14,10 +14,10 @@
 #include <linux/buffer_head.h>
 #include <linux/gfs2_ondisk.h>
 #include <linux/crc32.h>
-#include <asm/semaphore.h>
+#include <linux/lm_interface.h>
+#include <linux/delay.h>
 
 #include "gfs2.h"
-#include "lm_interface.h"
 #include "incore.h"
 #include "bmap.h"
 #include "glock.h"
@@ -48,8 +48,7 @@ unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct,
 	unsigned int first, second;
 
 	blks = 1;
-	first = (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_log_descriptor)) /
-		ssize;
+	first = (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_log_descriptor)) / ssize;
 
 	if (nstruct > first) {
 		second = (sdp->sd_sb.sb_bsize -
@@ -60,12 +59,97 @@ unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct,
 	return blks;
 }
 
-void gfs2_ail1_start(struct gfs2_sbd *sdp, int flags)
+/**
+ * gfs2_ail1_start_one - Start I/O on a part of the AIL
+ * @sdp: the filesystem
+ * @tr: the part of the AIL
+ *
+ */
+
+static void gfs2_ail1_start_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai)
+{
+	struct gfs2_bufdata *bd, *s;
+	struct buffer_head *bh;
+	int retry;
+
+	BUG_ON(!spin_is_locked(&sdp->sd_log_lock));
+
+	do {
+		retry = 0;
+
+		list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list,
+						 bd_ail_st_list) {
+			bh = bd->bd_bh;
+
+			gfs2_assert(sdp, bd->bd_ail == ai);
+
+			if (!buffer_busy(bh)) {
+				if (!buffer_uptodate(bh)) {
+					gfs2_log_unlock(sdp);
+					gfs2_io_error_bh(sdp, bh);
+					gfs2_log_lock(sdp);
+				}
+				list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list);
+				continue;
+			}
+
+			if (!buffer_dirty(bh))
+				continue;
+
+			list_move(&bd->bd_ail_st_list, &ai->ai_ail1_list);
+
+			gfs2_log_unlock(sdp);
+			wait_on_buffer(bh);
+			ll_rw_block(WRITE, 1, &bh);
+			gfs2_log_lock(sdp);
+
+			retry = 1;
+			break;
+		}
+	} while (retry);
+}
+
+/**
+ * gfs2_ail1_empty_one - Check whether or not a trans in the AIL has been synced
+ * @sdp: the filesystem
+ * @ai: the AIL entry
+ *
+ */
+
+static int gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai, int flags)
+{
+	struct gfs2_bufdata *bd, *s;
+	struct buffer_head *bh;
+
+	list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list,
+					 bd_ail_st_list) {
+		bh = bd->bd_bh;
+
+		gfs2_assert(sdp, bd->bd_ail == ai);
+
+		if (buffer_busy(bh)) {
+			if (flags & DIO_ALL)
+				continue;
+			else
+				break;
+		}
+
+		if (!buffer_uptodate(bh))
+			gfs2_io_error_bh(sdp, bh);
+
+		list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list);
+	}
+
+	return list_empty(&ai->ai_ail1_list);
+}
+
+static 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;
+	u64 sync_gen;
+	struct list_head *first;
+	struct gfs2_ail *first_ai, *ai, *tmp;
+	int done = 0;
 
 	gfs2_log_lock(sdp);
 	if (list_empty(head)) {
@@ -77,27 +161,25 @@ void gfs2_ail1_start(struct gfs2_sbd *sdp, int flags)
 	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);
+	gfs2_ail1_start_one(sdp, first_ai); /* This may drop log lock */
 
 	if (flags & DIO_ALL)
 		first = NULL;
 
-	for (;;) {
+	while(!done) {
 		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);
+		done = 1;
+		list_for_each_entry_safe_reverse(ai, tmp, head, ai_list) {
 			if (ai->ai_sync_gen >= sync_gen)
 				continue;
 			ai->ai_sync_gen = sync_gen;
-			gfs2_ail1_start_one(sdp, ai);
+			gfs2_ail1_start_one(sdp, ai); /* This may drop log lock */
+			done = 0;
 			break;
 		}
-
-		if (tmp == head)
-			break;
 	}
 
 	gfs2_log_unlock(sdp);
@@ -124,6 +206,31 @@ int gfs2_ail1_empty(struct gfs2_sbd *sdp, int flags)
 	return ret;
 }
 
+
+/**
+ * gfs2_ail2_empty_one - Check whether or not a trans in the AIL has been synced
+ * @sdp: the filesystem
+ * @ai: the AIL entry
+ *
+ */
+
+static void gfs2_ail2_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai)
+{
+	struct list_head *head = &ai->ai_ail2_list;
+	struct gfs2_bufdata *bd;
+
+	while (!list_empty(head)) {
+		bd = list_entry(head->prev, struct gfs2_bufdata,
+				bd_ail_st_list);
+		gfs2_assert(sdp, bd->bd_ail == ai);
+		bd->bd_ail = NULL;
+		list_del(&bd->bd_ail_st_list);
+		list_del(&bd->bd_ail_gl_list);
+		atomic_dec(&bd->bd_gl->gl_ail_count);
+		brelse(bd->bd_bh);
+	}
+}
+
 static void ail2_empty(struct gfs2_sbd *sdp, unsigned int new_tail)
 {
 	struct gfs2_ail *ai, *safe;
@@ -155,6 +262,12 @@ static void ail2_empty(struct gfs2_sbd *sdp, unsigned int new_tail)
  * @sdp: The GFS2 superblock
  * @blks: The number of blocks to reserve
  *
+ * Note that we never give out the last 6 blocks of the journal. Thats
+ * due to the fact that there is are a small number of header blocks
+ * associated with each log flush. The exact number can't be known until
+ * flush time, so we ensure that we have just enough free blocks at all
+ * times to avoid running out during a log flush.
+ *
  * Returns: errno
  */
 
@@ -168,7 +281,7 @@ int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks)
 
 	mutex_lock(&sdp->sd_log_reserve_mutex);
 	gfs2_log_lock(sdp);
-	while(sdp->sd_log_blks_free <= blks) {
+	while(sdp->sd_log_blks_free <= (blks + 6)) {
 		gfs2_log_unlock(sdp);
 		gfs2_ail1_empty(sdp, 0);
 		gfs2_log_flush(sdp, NULL);
@@ -178,7 +291,6 @@ int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks)
 		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);
 
@@ -199,24 +311,26 @@ 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)
+static u64 log_bmap(struct gfs2_sbd *sdp, unsigned int lbn)
 {
-	int new = 0;
-	uint64_t dbn;
+	struct inode *inode = sdp->sd_jdesc->jd_inode;
 	int error;
+	struct buffer_head bh_map = { .b_state = 0, .b_blocknr = 0 };
 
-	error = gfs2_block_map(sdp->sd_jdesc->jd_inode->u.generic_ip,
-			       lbn, &new, &dbn, NULL);
-	gfs2_assert_withdraw(sdp, !error && dbn);
+	bh_map.b_size = 1 << inode->i_blkbits;
+	error = gfs2_block_map(inode, lbn, 0, &bh_map);
+	if (error || !bh_map.b_blocknr)
+		printk(KERN_INFO "error=%d, dbn=%llu lbn=%u", error,
+		       (unsigned long long)bh_map.b_blocknr, lbn);
+	gfs2_assert_withdraw(sdp, !error && bh_map.b_blocknr);
 
-	return dbn;
+	return bh_map.b_blocknr;
 }
 
 /**
@@ -231,8 +345,7 @@ static uint64_t log_bmap(struct gfs2_sbd *sdp, unsigned int lbn)
  * Returns: the distance in blocks
  */
 
-static inline unsigned int log_distance(struct gfs2_sbd *sdp,
-					unsigned int newer,
+static inline unsigned int log_distance(struct gfs2_sbd *sdp, unsigned int newer,
 					unsigned int older)
 {
 	int dist;
@@ -251,11 +364,10 @@ static unsigned int current_tail(struct gfs2_sbd *sdp)
 
 	gfs2_log_lock(sdp);
 
-	if (list_empty(&sdp->sd_ail1_list))
+	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);
+	} else {
+		ai = list_entry(sdp->sd_ail1_list.prev, struct gfs2_ail, ai_list);
 		tail = ai->ai_first;
 	}
 
@@ -267,8 +379,7 @@ static unsigned int current_tail(struct gfs2_sbd *sdp)
 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);
+		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;
@@ -285,7 +396,7 @@ static inline void log_incr_head(struct gfs2_sbd *sdp)
 
 struct buffer_head *gfs2_log_get_buf(struct gfs2_sbd *sdp)
 {
-	uint64_t blkno = log_bmap(sdp, sdp->sd_log_flush_head);
+	u64 blkno = log_bmap(sdp, sdp->sd_log_flush_head);
 	struct gfs2_log_buf *lb;
 	struct buffer_head *bh;
 
@@ -315,7 +426,7 @@ 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)
 {
-	uint64_t blkno = log_bmap(sdp, sdp->sd_log_flush_head);
+	u64 blkno = log_bmap(sdp, sdp->sd_log_flush_head);
 	struct gfs2_log_buf *lb;
 	struct buffer_head *bh;
 
@@ -343,10 +454,8 @@ static void log_pull_tail(struct gfs2_sbd *sdp, unsigned int new_tail, int pull)
 	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);
+	sdp->sd_log_blks_free += dist - (pull ? 1 : 0);
+	gfs2_assert_withdraw(sdp, sdp->sd_log_blks_free <= sdp->sd_jdesc->jd_blocks);
 	gfs2_log_unlock(sdp);
 
 	sdp->sd_log_tail = new_tail;
@@ -359,15 +468,13 @@ static void log_pull_tail(struct gfs2_sbd *sdp, unsigned int new_tail, int pull)
  * Returns: the initialized log buffer descriptor
  */
 
-static void log_write_header(struct gfs2_sbd *sdp, uint32_t flags, int pull)
+static void log_write_header(struct gfs2_sbd *sdp, u32 flags, int pull)
 {
-	uint64_t blkno = log_bmap(sdp, sdp->sd_log_flush_head);
+	u64 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); */
+	u32 hash;
 
 	bh = sb_getblk(sdp->sd_vfs, blkno);
 	lock_buffer(bh);
@@ -384,10 +491,10 @@ static void log_write_header(struct gfs2_sbd *sdp, uint32_t flags, int pull)
 	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 = be64_to_cpu(sdp->sd_log_sequence++);
-	lh->lh_flags = be32_to_cpu(flags);
-	lh->lh_tail = be32_to_cpu(tail);
-	lh->lh_blkno = be32_to_cpu(sdp->sd_log_flush_head);
+	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);
 
@@ -403,8 +510,6 @@ static void log_write_header(struct gfs2_sbd *sdp, uint32_t flags, int 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)
@@ -412,11 +517,6 @@ 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;
-	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);
 
 	while (!list_empty(head)) {
 		lb = list_entry(head->next, struct gfs2_log_buf, lb_list);
@@ -465,8 +565,7 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl)
 	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_buf == sdp->sd_log_commited_buf);
 	gfs2_assert_withdraw(sdp,
 			sdp->sd_log_num_revoke == sdp->sd_log_commited_revoke);
 
@@ -480,16 +579,15 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl)
 	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_reserved =
-		sdp->sd_log_commited_buf =
-		sdp->sd_log_num_hdrs = 
-		sdp->sd_log_commited_revoke = 0;
 
 	gfs2_log_lock(sdp);
+	sdp->sd_log_head = sdp->sd_log_flush_head;
+	sdp->sd_log_blks_free -= sdp->sd_log_num_hdrs;
+	sdp->sd_log_blks_reserved = 0;
+	sdp->sd_log_commited_buf = 0;
+	sdp->sd_log_num_hdrs = 0;
+	sdp->sd_log_commited_revoke = 0;
+
 	if (!list_empty(&ai->ai_ail1_list)) {
 		list_add(&ai->ai_list, &sdp->sd_ail1_list);
 		ai = NULL;
@@ -504,7 +602,7 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl)
 
 static void log_refund(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 {
-	unsigned int reserved = 1;
+	unsigned int reserved = 0;
 	unsigned int old;
 
 	gfs2_log_lock(sdp);
@@ -515,11 +613,12 @@ static void log_refund(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 	gfs2_assert_withdraw(sdp, ((int)sdp->sd_log_commited_revoke) >= 0);
 
 	if (sdp->sd_log_commited_buf)
-		reserved += 1 + sdp->sd_log_commited_buf +
-			    sdp->sd_log_commited_buf/503;
+		reserved += sdp->sd_log_commited_buf;
 	if (sdp->sd_log_commited_revoke)
 		reserved += gfs2_struct2blk(sdp, sdp->sd_log_commited_revoke,
-					    sizeof(uint64_t));
+					    sizeof(u64));
+	if (reserved)
+		reserved++;
 
 	old = sdp->sd_log_blks_free;
 	sdp->sd_log_blks_free += tr->tr_reserved -
@@ -527,7 +626,8 @@ static void log_refund(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 
 	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_blks_free <= sdp->sd_jdesc->jd_blocks +
+			     sdp->sd_log_num_hdrs);
 
 	sdp->sd_log_blks_reserved = reserved;
 
@@ -551,11 +651,9 @@ void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 	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);
+	if (sdp->sd_log_num_buf > gfs2_tune_get(sdp, gt_incore_log_blocks))
+		wake_up_process(sdp->sd_logd_process);
+	gfs2_log_unlock(sdp);
 }
 
 /**
@@ -575,6 +673,7 @@ void gfs2_log_shutdown(struct gfs2_sbd *sdp)
 	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;
@@ -582,10 +681,9 @@ void gfs2_log_shutdown(struct gfs2_sbd *sdp)
 
 	log_write_header(sdp, GFS2_LOG_HEAD_UNMOUNT, 0);
 
-	gfs2_assert_withdraw(sdp, sdp->sd_log_blks_free ==
-			     sdp->sd_jdesc->jd_blocks);
-	gfs2_assert_withdraw(sdp, sdp->sd_log_head == sdp->sd_log_tail);
-	gfs2_assert_withdraw(sdp, list_empty(&sdp->sd_ail2_list));
+	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;
@@ -593,3 +691,21 @@ void gfs2_log_shutdown(struct gfs2_sbd *sdp)
 	up_write(&sdp->sd_log_flush_lock);
 }
 
+
+/**
+ * gfs2_meta_syncfs - sync all the buffers in a filesystem
+ * @sdp: the filesystem
+ *
+ */
+
+void gfs2_meta_syncfs(struct gfs2_sbd *sdp)
+{
+	gfs2_log_flush(sdp, NULL);
+	for (;;) {
+		gfs2_ail1_start(sdp, DIO_ALL);
+		if (gfs2_ail1_empty(sdp, DIO_ALL))
+			break;
+		msleep(10);
+	}
+}
+