From: Vyacheslav Dubeyko <slava@xxxxxxxxxxx>
Subject: [PATCH 11/24] nilfs2: implement recovery subsystem debugging output
This patch adds debugging output by means of nilfs2_debug() method
into modules that are grouped by recovery subsystem debugging output
option (CONFIG_NILFS2_DEBUG_RECOVERY_SUBSYSTEM).
Signed-off-by: Vyacheslav Dubeyko <slava@xxxxxxxxxxx>
CC: Ryusuke Konishi <konishi.ryusuke@xxxxxxxxxxxxx>
---
fs/nilfs2/recovery.c | 63 ++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 63 insertions(+)
diff --git a/fs/nilfs2/recovery.c b/fs/nilfs2/recovery.c
index ff00a0b..1099ba1 100644
--- a/fs/nilfs2/recovery.c
+++ b/fs/nilfs2/recovery.c
@@ -110,6 +110,11 @@ static int nilfs_compute_checksum(struct the_nilfs *nilfs,
unsigned long size;
u32 crc;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, bhs %p, sum %p, offset %lu"
+ "check_bytes %llu, start %lu, nblock %lu\n",
+ nilfs, bhs, sum, offset, check_bytes, start, nblock);
+
BUG_ON(offset >= blocksize);
check_bytes -= offset;
size = min_t(u64, check_bytes, blocksize - offset);
@@ -147,6 +152,10 @@ int nilfs_read_super_root_block(struct the_nilfs *nilfs, sector_t sr_block,
u32 crc;
int ret;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, sr_block %lu, pbh %p, check %d\n",
+ nilfs, sr_block, pbh, check);
+
*pbh = NULL;
bh_sr = __bread(nilfs->ns_bdev, sr_block, nilfs->ns_blocksize);
if (unlikely(!bh_sr)) {
@@ -195,6 +204,10 @@ nilfs_read_log_header(struct the_nilfs *nilfs, sector_t start_blocknr,
{
struct buffer_head *bh_sum;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, start_blocknr %lu, sum %p\n",
+ nilfs, start_blocknr, sum);
+
bh_sum = __bread(nilfs->ns_bdev, start_blocknr, nilfs->ns_blocksize);
if (bh_sum)
*sum = (struct nilfs_segment_summary *)bh_sum->b_data;
@@ -216,6 +229,10 @@ static int nilfs_validate_log(struct the_nilfs *nilfs, u64 seg_seq,
u32 crc;
int ret;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, seg_seq %llu, bh_sum %p, sum %p\n",
+ nilfs, seg_seq, bh_sum, sum);
+
ret = NILFS_SEG_FAIL_MAGIC;
if (le32_to_cpu(sum->ss_magic) != NILFS_SEGSUM_MAGIC)
goto out;
@@ -258,6 +275,10 @@ static void *nilfs_read_summary_info(struct the_nilfs *nilfs,
void *ptr;
sector_t blocknr;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, pbh %p, offset %u, bytes %u\n",
+ nilfs, pbh, *offset, bytes);
+
BUG_ON((*pbh)->b_size < *offset);
if (bytes > (*pbh)->b_size - *offset) {
blocknr = (*pbh)->b_blocknr;
@@ -270,6 +291,9 @@ static void *nilfs_read_summary_info(struct the_nilfs *nilfs,
}
ptr = (*pbh)->b_data + *offset;
*offset += bytes;
+
+ nilfs2_debug(DBG_RECOVERY, "calculated offset %u\n", *offset);
+
return ptr;
}
@@ -289,6 +313,10 @@ static void nilfs_skip_summary_info(struct the_nilfs *nilfs,
unsigned int rest_item_in_current_block
= ((*pbh)->b_size - *offset) / bytes;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, pbh %p, offset %u, bytes %u, count %lu\n",
+ nilfs, pbh, *offset, bytes, count);
+
if (count <= rest_item_in_current_block) {
*offset += bytes * count;
} else {
@@ -304,6 +332,8 @@ static void nilfs_skip_summary_info(struct the_nilfs *nilfs,
*pbh = __bread(nilfs->ns_bdev, blocknr + bcnt,
nilfs->ns_blocksize);
}
+
+ nilfs2_debug(DBG_RECOVERY, "calculated offset %u\n", *offset);
}
/**
@@ -324,6 +354,10 @@ static int nilfs_scan_dsync_log(struct the_nilfs *nilfs, sector_t start_blocknr,
ino_t ino;
int err = -EIO;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, start_blocknr %lu, sum %p, head %p\n",
+ nilfs, start_blocknr, sum, head);
+
nfinfo = le32_to_cpu(sum->ss_nfinfo);
if (!nfinfo)
return 0;
@@ -404,6 +438,9 @@ static int nilfs_segment_list_add(struct list_head *head, __u64 segnum)
{
struct nilfs_segment_entry *ent = kmalloc(sizeof(*ent), GFP_NOFS);
+ nilfs2_debug(DBG_RECOVERY,
+ "head %p, segnum %llu\n", head, segnum);
+
if (unlikely(!ent))
return -ENOMEM;
@@ -435,6 +472,9 @@ static int nilfs_prepare_segment_for_recovery(struct the_nilfs *nilfs,
int err;
int i;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, sb %p, ri %p\n", nilfs, sb, ri);
+
segnum[0] = nilfs->ns_segnum;
segnum[1] = nilfs->ns_nextnum;
segnum[2] = ri->ri_segnum;
@@ -489,6 +529,10 @@ static int nilfs_recovery_copy_block(struct the_nilfs *nilfs,
struct buffer_head *bh_org;
void *kaddr;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, rb %p, i_ino %lu\n",
+ nilfs, rb, page->mapping->host->i_ino);
+
bh_org = __bread(nilfs->ns_bdev, rb->blocknr, nilfs->ns_blocksize);
if (unlikely(!bh_org))
return -EIO;
@@ -513,6 +557,11 @@ static int nilfs_recover_dsync_blocks(struct the_nilfs *nilfs,
loff_t pos;
int err = 0, err2 = 0;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, sb %p, root %p, "
+ "head %p, nr_salvaged_blocks %p\n",
+ nilfs, sb, root, head, nr_salvaged_blocks);
+
list_for_each_entry_safe(rb, n, head, list) {
inode = nilfs_iget(sb, root, rb->ino);
if (IS_ERR(inode)) {
@@ -598,6 +647,10 @@ static int nilfs_do_roll_forward(struct the_nilfs *nilfs,
};
int state = RF_INIT_ST;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, sb %p, root %p, ri %p\n",
+ nilfs, sb, root, ri);
+
pseg_start = ri->ri_lsegs_start;
seg_seq = ri->ri_lsegs_start_seq;
segnum = nilfs_get_segnum_of_block(nilfs, pseg_start);
@@ -706,6 +759,9 @@ static void nilfs_finish_roll_forward(struct the_nilfs *nilfs,
struct buffer_head *bh;
int err;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, ri %p\n", nilfs, ri);
+
if (nilfs_get_segnum_of_block(nilfs, ri->ri_lsegs_start) !=
nilfs_get_segnum_of_block(nilfs, ri->ri_super_root))
return;
@@ -748,6 +804,10 @@ int nilfs_salvage_orphan_logs(struct the_nilfs *nilfs,
struct nilfs_root *root;
int err;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, sb %p, ri %p\n",
+ nilfs, sb, ri);
+
if (ri->ri_lsegs_start == 0 || ri->ri_lsegs_end == 0)
return 0;
@@ -827,6 +887,9 @@ int nilfs_search_super_root(struct the_nilfs *nilfs,
int empty_seg = 0, scan_newer = 0;
int ret;
+ nilfs2_debug(DBG_RECOVERY,
+ "nilfs %p, ri %p\n", nilfs, ri);
+
pseg_start = nilfs->ns_last_pseg;
seg_seq = nilfs->ns_last_seq;
cno = nilfs->ns_last_cno;
--
1.7.9.5
--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
