--- /dev/null
+Index: linux-2.6.18.i686/fs/ext4/mballoc.c
+===================================================================
+--- linux-2.6.18.i686.orig/fs/ext4/mballoc.c
++++ linux-2.6.18.i686/fs/ext4/mballoc.c
+@@ -660,7 +660,7 @@ static void ext4_mb_mark_free_simple(str
+ }
+ }
+
+-static void ext4_mb_generate_buddy(struct super_block *sb,
++static int ext4_mb_generate_buddy(struct super_block *sb,
+ void *buddy, void *bitmap, ext4_group_t group)
+ {
+ struct ext4_group_info *grp = ext4_get_group_info(sb, group);
+@@ -692,14 +692,14 @@ static void ext4_mb_generate_buddy(struc
+ grp->bb_fragments = fragments;
+
+ if (free != grp->bb_free) {
+- ext4_error(sb, __func__,
+- "EXT4-fs: group %lu: %u blocks in bitmap, %u in gd\n",
+- group, free, grp->bb_free);
+- /*
+- * If we intent to continue, we consider group descritor
+- * corrupt and update bb_free using bitmap value
+- */
+- grp->bb_free = free;
++ struct ext4_group_desc *gdp;
++ gdp = ext4_get_group_desc (sb, group, NULL);
++ ext4_error(sb, __FUNCTION__,
++ "group %lu: %u blocks in bitmap, %u in bb, "
++ "%u in gd, %lu pa's\n", group, free, grp->bb_free,
++ le16_to_cpu(gdp->bg_free_blocks_count),
++ grp->bb_prealloc_nr);
++ return -EIO;
+ }
+
+ clear_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &(grp->bb_state));
+@@ -709,6 +709,8 @@ static void ext4_mb_generate_buddy(struc
+ EXT4_SB(sb)->s_mb_buddies_generated++;
+ EXT4_SB(sb)->s_mb_generation_time += period;
+ spin_unlock(&EXT4_SB(sb)->s_bal_lock);
++
++ return 0;
+ }
+
+ /* The buddy information is attached the buddy cache inode
+@@ -814,7 +816,7 @@ static int ext4_mb_init_cache(struct pag
+
+ err = 0;
+ first_block = page->index * blocks_per_page;
+- for (i = 0; i < blocks_per_page; i++) {
++ for (i = 0; i < blocks_per_page && err == 0; i++) {
+ int group;
+ struct ext4_group_info *grinfo;
+
+@@ -848,7 +850,7 @@ static int ext4_mb_init_cache(struct pag
+ /*
+ * incore got set to the group block bitmap below
+ */
+- ext4_mb_generate_buddy(sb, data, incore, group);
++ err = ext4_mb_generate_buddy(sb, data, incore, group);
+ incore = NULL;
+ } else {
+ /* this is block of bitmap */
+@@ -861,7 +863,7 @@ static int ext4_mb_init_cache(struct pag
+ memcpy(data, bitmap, blocksize);
+
+ /* mark all preallocated blks used in in-core bitmap */
+- ext4_mb_generate_from_pa(sb, data, group);
++ err = ext4_mb_generate_from_pa(sb, data, group);
+ ext4_unlock_group(sb, group);
+
+ /* set incore so that the buddy information can be
+@@ -870,6 +872,7 @@ static int ext4_mb_init_cache(struct pag
+ incore = data;
+ }
+ }
++ if (likely(err == 0))
+ SetPageUptodate(page);
+
+ out:
+@@ -1964,7 +1967,10 @@ static int ext4_mb_seq_history_show(stru
+ hs->result.fe_start, hs->result.fe_len);
+ seq_printf(seq, "%-5u %-8u %-23s free\n",
+ hs->pid, hs->ino, buf2);
++ } else {
++ seq_printf(seq, "unknown op %d\n", hs->op);
+ }
++
+ return 0;
+ }
+
+@@ -2092,9 +2098,11 @@ static void *ext4_mb_seq_groups_next(str
+ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v)
+ {
+ struct super_block *sb = seq->private;
++ struct ext4_group_desc *gdp;
+ long group = (long) v;
+ int i;
+ int err;
++ int free = 0;
+ struct ext4_buddy e4b;
+ struct sg {
+ struct ext4_group_info info;
+@@ -2103,10 +2111,10 @@ static int ext4_mb_seq_groups_show(struc
+
+ group--;
+ if (group == 0)
+- seq_printf(seq, "#%-5s: %-5s %-5s %-5s "
++ seq_printf(seq, "#%-5s: %-5s %-5s %-5s %-5s %-5s"
+ "[ %-5s %-5s %-5s %-5s %-5s %-5s %-5s "
+ "%-5s %-5s %-5s %-5s %-5s %-5s %-5s ]\n",
+- "group", "free", "frags", "first",
++ "group", "free", "frags", "first", "first", "pa",
+ "2^0", "2^1", "2^2", "2^3", "2^4", "2^5", "2^6",
+ "2^7", "2^8", "2^9", "2^10", "2^11", "2^12", "2^13");
+
+@@ -2117,13 +2125,20 @@ static int ext4_mb_seq_groups_show(struc
+ seq_printf(seq, "#%-5lu: I/O error\n", group);
+ return 0;
+ }
++
++ gdp = ext4_get_group_desc(sb, group, NULL);
++ if (gdp != NULL)
++ free = le16_to_cpu(gdp->bg_free_blocks_count);
++
+ ext4_lock_group(sb, group);
+ memcpy(&sg, ext4_get_group_info(sb, group), i);
+ ext4_unlock_group(sb, group);
+ ext4_mb_release_desc(&e4b);
+
+- seq_printf(seq, "#%-5lu: %-5u %-5u %-5u [", group, sg.info.bb_free,
+- sg.info.bb_fragments, sg.info.bb_first_free);
++ seq_printf(seq, "#%-5lu: %-5u %-5u %-5u %-5u %-5lu [", group,
++ sg.info.bb_free, free,
++ sg.info.bb_fragments, sg.info.bb_first_free,
++ sg.info.bb_prealloc_nr);
+ for (i = 0; i <= 13; i++)
+ seq_printf(seq, " %-5u", i <= sb->s_blocksize_bits + 1 ?
+ sg.info.bb_counters[i] : 0);
+@@ -2226,6 +2241,7 @@ ext4_mb_store_history(struct ext4_alloca
+ h.tail = ac->ac_tail;
+ h.buddy = ac->ac_buddy;
+ h.merged = 0;
++ h.cr = ac->ac_criteria;
+ if (ac->ac_op == EXT4_MB_HISTORY_ALLOC) {
+ if (ac->ac_g_ex.fe_start == ac->ac_b_ex.fe_start &&
+ ac->ac_g_ex.fe_group == ac->ac_b_ex.fe_group)
+@@ -3531,22 +3547,66 @@ ext4_mb_use_preallocated(struct ext4_all
+ }
+
+ /*
++ * check free blocks in bitmap match free block in group descriptor
++ * do this before taking preallocated blocks into account to be able
++ * to detect on-disk corruptions
++ */
++int ext4_mb_check_ondisk_bitmap(struct super_block *sb, void *bitmap,
++ struct ext4_group_desc *gdp, int group)
++{
++ unsigned short max = EXT4_BLOCKS_PER_GROUP(sb);
++ unsigned short i, first, free = 0;
++
++ i = mb_find_next_zero_bit(bitmap, max, 0);
++
++ while (i < max) {
++ first = i;
++ i = find_next_bit(bitmap, max, i);
++ if (i > max)
++ i = max;
++ free += i - first;
++ if (i < max)
++ i = mb_find_next_zero_bit(bitmap, max, i);
++ }
++
++ if (free != le16_to_cpu(gdp->bg_free_blocks_count)) {
++ ext4_error(sb, __FUNCTION__, "on-disk bitmap for group %d"
++ "corrupted: %u blocks free in bitmap, %u - in gd\n",
++ group, free, le16_to_cpu(gdp->bg_free_blocks_count));
++ return -EIO;
++ }
++ return 0;
++}
++
++/*
+ * the function goes through all preallocation in this group and marks them
+ * used in in-core bitmap. buddy must be generated from this bitmap
+ * Need to be called with ext4 group lock (ext4_lock_group)
+ */
+-static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap,
++static int ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap,
+ ext4_group_t group)
+ {
+ struct ext4_group_info *grp = ext4_get_group_info(sb, group);
+ struct ext4_prealloc_space *pa;
++ struct ext4_group_desc *gdp;
+ struct list_head *cur;
+ ext4_group_t groupnr;
+ ext4_grpblk_t start;
+ int preallocated = 0;
+ int count = 0;
++ int skip = 0;
++ int err;
+ int len;
+
++ gdp = ext4_get_group_desc (sb, group, NULL);
++ if (gdp == NULL)
++ return -EIO;
++
++ /* before applying preallocations, check bitmap consistency */
++ err = ext4_mb_check_ondisk_bitmap(sb, bitmap, gdp, group);
++ if (err)
++ return err;
++
+ /* all form of preallocation discards first load group,
+ * so the only competing code is preallocation use.
+ * we don't need any locking here
+@@ -3562,15 +3622,24 @@ static void ext4_mb_generate_from_pa(str
+ &groupnr, &start);
+ len = pa->pa_len;
+ spin_unlock(&pa->pa_lock);
+- if (unlikely(len == 0))
++ if (unlikely(len == 0)) {
++ skip++;
+ continue;
++ }
+ BUG_ON(groupnr != group);
+ mb_set_bits(sb_bgl_lock(EXT4_SB(sb), group),
+ bitmap, start, len);
+ preallocated += len;
+ count++;
+ }
++ if (count + skip != grp->bb_prealloc_nr) {
++ ext4_error(sb, __FUNCTION__, "lost preallocations: "
++ "count %d, bb_prealloc_nr %lu, skip %d\n",
++ count, grp->bb_prealloc_nr, skip);
++ return -EIO;
++ }
+ mb_debug("prellocated %u for group %lu\n", preallocated, group);
++ return 0;
+ }
+
+ static void ext4_mb_pa_callback(struct rcu_head *head)
+@@ -3621,6 +3690,7 @@ static void ext4_mb_put_pa(struct ext4_a
+ */
+ ext4_lock_group(sb, grp);
+ list_del(&pa->pa_group_list);
++ ext4_get_group_info(sb, grp)->bb_prealloc_nr--;
+ ext4_unlock_group(sb, grp);
+
+ spin_lock(pa->pa_obj_lock);
+@@ -3709,6 +3779,7 @@ ext4_mb_new_inode_pa(struct ext4_allocat
+
+ ext4_lock_group(sb, ac->ac_b_ex.fe_group);
+ list_add(&pa->pa_group_list, &grp->bb_prealloc_list);
++ grp->bb_prealloc_nr++;
+ ext4_unlock_group(sb, ac->ac_b_ex.fe_group);
+
+ spin_lock(pa->pa_obj_lock);
+@@ -3768,6 +3839,7 @@ ext4_mb_new_group_pa(struct ext4_allocat
+
+ ext4_lock_group(sb, ac->ac_b_ex.fe_group);
+ list_add(&pa->pa_group_list, &grp->bb_prealloc_list);
++ grp->bb_prealloc_nr++;
+ ext4_unlock_group(sb, ac->ac_b_ex.fe_group);
+
+ /*
+@@ -3820,6 +3892,7 @@ ext4_mb_release_inode_pa(struct ext4_bud
+ ac->ac_sb = sb;
+ ac->ac_inode = pa->pa_inode;
+ ac->ac_op = EXT4_MB_HISTORY_DISCARD;
++ ac->ac_o_ex.fe_len = 1;
+ }
+
+ while (bit < end) {
+@@ -3964,6 +4037,8 @@ repeat:
+
+ spin_unlock(&pa->pa_lock);
+
++ BUG_ON(grp->bb_prealloc_nr == 0);
++ grp->bb_prealloc_nr--;
+ list_del(&pa->pa_group_list);
+ list_add(&pa->u.pa_tmp_list, &list);
+ }
+@@ -4099,7 +4174,7 @@ repeat:
+ if (err) {
+ ext4_error(sb, __func__, "Error in loading buddy "
+ "information for %lu\n", group);
+- continue;
++ return;
+ }
+
+ bitmap_bh = ext4_read_block_bitmap(sb, group);
+@@ -4111,6 +4186,8 @@ repeat:
+ }
+
+ ext4_lock_group(sb, group);
++ BUG_ON(e4b.bd_info->bb_prealloc_nr == 0);
++ e4b.bd_info->bb_prealloc_nr--;
+ list_del(&pa->pa_group_list);
+ ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa, ac);
+ ext4_unlock_group(sb, group);
+Index: linux-2.6.18.i686/fs/ext4/mballoc.h
+===================================================================
+--- linux-2.6.18.i686.orig/fs/ext4/mballoc.h
++++ linux-2.6.18.i686/fs/ext4/mballoc.h
+@@ -119,6 +119,7 @@ struct ext4_group_info {
+ unsigned short bb_free;
+ unsigned short bb_fragments;
+ struct list_head bb_prealloc_list;
++ unsigned long bb_prealloc_nr;
+ #ifdef DOUBLE_CHECK
+ void *bb_bitmap;
+ #endif
+@@ -228,7 +229,7 @@ struct ext4_mb_history {
+ __u16 tail; /* what tail broke some buddy */
+ __u16 buddy; /* buddy the tail ^^^ broke */
+ __u16 flags;
+- __u8 cr:3; /* which phase the result extent was found at */
++ __u8 cr:8; /* which phase the result extent was found at */
+ __u8 op:4;
+ __u8 merged:1;
+ };
+@@ -259,7 +260,7 @@ static void ext4_mb_store_history(struct
+
+ struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t);
+
+-static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap,
++static int ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap,
+ ext4_group_t group);
+ static void ext4_mb_poll_new_transaction(struct super_block *, handle_t *);
+ static void ext4_mb_free_committed_blocks(struct super_block *);
git://git.whamcloud.com / fs / lustre-release.git / blobdiff