With this change new blocks added during resize
are marked as free in the block bitmap and the
group is flagged with EXT4_GROUP_INFO_NEED_INIT_BIT
flag. This make sure when mballoc tries to allocate
blocks from the new group we would reload the
buddy information using the bitmap present in the disk.
Signed-off-by: Aneesh Kumar K.V <[email protected]>
---
fs/ext4/balloc.c | 136 ++++++++++++------------------------------------------
fs/ext4/ext4.h | 5 +-
fs/ext4/resize.c | 11 +----
3 files changed, 34 insertions(+), 118 deletions(-)
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index 4397274..d598fd6 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -20,6 +20,7 @@
#include "ext4.h"
#include "ext4_jbd2.h"
#include "group.h"
+#include "mballoc.h"
/*
* balloc.c contains the blocks allocation and deallocation routines
@@ -350,62 +351,43 @@ ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group)
}
/**
- * ext4_free_blocks_sb() -- Free given blocks and update quota
+ * ext4_add_groupblocks() -- Add given blocks to an existing group
* @handle: handle to this transaction
* @sb: super block
- * @block: start physcial block to free
+ * @block: start physcial block to add to the block group
* @count: number of blocks to free
- * @pdquot_freed_blocks: pointer to quota
*
- * XXX This function is only used by the on-line resizing code, which
- * should probably be fixed up to call the mballoc variant. There
- * this needs to be cleaned up later; in fact, I'm not convinced this
- * is 100% correct in the face of the mballoc code. The online resizing
- * code needs to be fixed up to more tightly (and correctly) interlock
- * with the mballoc code.
+ * This marks the blocks as free in the bitmap. We ask the
+ * mballoc to reload the buddy after this by setting group
+ * EXT4_GROUP_INFO_NEED_INIT_BIT flag
*/
-void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb,
- ext4_fsblk_t block, unsigned long count,
- unsigned long *pdquot_freed_blocks)
+void ext4_add_groupblocks(handle_t *handle, struct super_block *sb,
+ ext4_fsblk_t block, unsigned long count)
{
struct buffer_head *bitmap_bh = NULL;
struct buffer_head *gd_bh;
ext4_group_t block_group;
ext4_grpblk_t bit;
unsigned long i;
- unsigned long overflow;
struct ext4_group_desc *desc;
struct ext4_super_block *es;
struct ext4_sb_info *sbi;
int err = 0, ret;
- ext4_grpblk_t group_freed;
+ ext4_grpblk_t blocks_freed;
+ struct ext4_group_info *grp;
- *pdquot_freed_blocks = 0;
sbi = EXT4_SB(sb);
es = sbi->s_es;
- if (block < le32_to_cpu(es->s_first_data_block) ||
- block + count < block ||
- block + count > ext4_blocks_count(es)) {
- ext4_error(sb, "ext4_free_blocks",
- "Freeing blocks not in datazone - "
- "block = %llu, count = %lu", block, count);
- goto error_return;
- }
-
- ext4_debug("freeing block(s) %llu-%llu\n", block, block + count - 1);
+ ext4_debug("Adding block(s) %llu-%llu\n", block, block + count - 1);
-do_more:
- overflow = 0;
ext4_get_group_no_and_offset(sb, block, &block_group, &bit);
/*
* Check to see if we are freeing blocks across a group
* boundary.
*/
if (bit + count > EXT4_BLOCKS_PER_GROUP(sb)) {
- overflow = bit + count - EXT4_BLOCKS_PER_GROUP(sb);
- count -= overflow;
+ goto error_return;
}
- brelse(bitmap_bh);
bitmap_bh = ext4_read_block_bitmap(sb, block_group);
if (!bitmap_bh)
goto error_return;
@@ -418,18 +400,17 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb,
in_range(block, ext4_inode_table(sb, desc), sbi->s_itb_per_group) ||
in_range(block + count - 1, ext4_inode_table(sb, desc),
sbi->s_itb_per_group)) {
- ext4_error(sb, "ext4_free_blocks",
- "Freeing blocks in system zones - "
+ ext4_error(sb, __func__,
+ "Adding blocks in system zones - "
"Block = %llu, count = %lu",
block, count);
goto error_return;
}
/*
- * We are about to start releasing blocks in the bitmap,
+ * We are about to add blocks to the bitmap,
* so we need undo access.
*/
- /* @@@ check errors */
BUFFER_TRACE(bitmap_bh, "getting undo access");
err = ext4_journal_get_undo_access(handle, bitmap_bh);
if (err)
@@ -445,87 +426,28 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb,
if (err)
goto error_return;
- jbd_lock_bh_state(bitmap_bh);
-
- for (i = 0, group_freed = 0; i < count; i++) {
- /*
- * An HJ special. This is expensive...
- */
-#ifdef CONFIG_JBD2_DEBUG
- jbd_unlock_bh_state(bitmap_bh);
- {
- struct buffer_head *debug_bh;
- debug_bh = sb_find_get_block(sb, block + i);
- if (debug_bh) {
- BUFFER_TRACE(debug_bh, "Deleted!");
- if (!bh2jh(bitmap_bh)->b_committed_data)
- BUFFER_TRACE(debug_bh,
- "No commited data in bitmap");
- BUFFER_TRACE2(debug_bh, bitmap_bh, "bitmap");
- __brelse(debug_bh);
- }
- }
- jbd_lock_bh_state(bitmap_bh);
-#endif
- if (need_resched()) {
- jbd_unlock_bh_state(bitmap_bh);
- cond_resched();
- jbd_lock_bh_state(bitmap_bh);
- }
- /* @@@ This prevents newly-allocated data from being
- * freed and then reallocated within the same
- * transaction.
- *
- * Ideally we would want to allow that to happen, but to
- * do so requires making jbd2_journal_forget() capable of
- * revoking the queued write of a data block, which
- * implies blocking on the journal lock. *forget()
- * cannot block due to truncate races.
- *
- * Eventually we can fix this by making jbd2_journal_forget()
- * return a status indicating whether or not it was able
- * to revoke the buffer. On successful revoke, it is
- * safe not to set the allocation bit in the committed
- * bitmap, because we know that there is no outstanding
- * activity on the buffer any more and so it is safe to
- * reallocate it.
- */
- BUFFER_TRACE(bitmap_bh, "set in b_committed_data");
- J_ASSERT_BH(bitmap_bh,
- bh2jh(bitmap_bh)->b_committed_data != NULL);
- ext4_set_bit_atomic(sb_bgl_lock(sbi, block_group), bit + i,
- bh2jh(bitmap_bh)->b_committed_data);
-
- /*
- * We clear the bit in the bitmap after setting the committed
- * data bit, because this is the reverse order to that which
- * the allocator uses.
- */
+ for (i = 0, blocks_freed = 0; i < count; i++) {
BUFFER_TRACE(bitmap_bh, "clear bit");
if (!ext4_clear_bit_atomic(sb_bgl_lock(sbi, block_group),
bit + i, bitmap_bh->b_data)) {
- jbd_unlock_bh_state(bitmap_bh);
ext4_error(sb, __func__,
"bit already cleared for block %llu",
(ext4_fsblk_t)(block + i));
- jbd_lock_bh_state(bitmap_bh);
BUFFER_TRACE(bitmap_bh, "bit already cleared");
} else {
- group_freed++;
+ blocks_freed++;
}
}
- jbd_unlock_bh_state(bitmap_bh);
-
spin_lock(sb_bgl_lock(sbi, block_group));
- le16_add_cpu(&desc->bg_free_blocks_count, group_freed);
+ le16_add_cpu(&desc->bg_free_blocks_count, blocks_freed);
desc->bg_checksum = ext4_group_desc_csum(sbi, block_group, desc);
spin_unlock(sb_bgl_lock(sbi, block_group));
- percpu_counter_add(&sbi->s_freeblocks_counter, count);
+ percpu_counter_add(&sbi->s_freeblocks_counter, blocks_freed);
if (sbi->s_log_groups_per_flex) {
ext4_group_t flex_group = ext4_flex_group(sbi, block_group);
spin_lock(sb_bgl_lock(sbi, flex_group));
- sbi->s_flex_groups[flex_group].free_blocks += count;
+ sbi->s_flex_groups[flex_group].free_blocks += blocks_freed;
spin_unlock(sb_bgl_lock(sbi, flex_group));
}
@@ -536,15 +458,17 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb,
/* And the group descriptor block */
BUFFER_TRACE(gd_bh, "dirtied group descriptor block");
ret = ext4_journal_dirty_metadata(handle, gd_bh);
- if (!err) err = ret;
- *pdquot_freed_blocks += group_freed;
-
- if (overflow && !err) {
- block += count;
- count = overflow;
- goto do_more;
- }
+ if (!err)
+ err = ret;
sb->s_dirt = 1;
+ /*
+ * request to reload the buddy with the
+ * new bitmap information
+ */
+ grp = ext4_get_group_info(sb, block_group);
+ set_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &(grp->bb_state));
+ ext4_mb_update_group_info(grp, blocks_freed);
+
error_return:
brelse(bitmap_bh);
ext4_std_error(sb, err);
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index fee616d..004dd25 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -1062,9 +1062,8 @@ extern int ext4_claim_free_blocks(struct ext4_sb_info *sbi, s64 nblocks);
extern int ext4_has_free_blocks(struct ext4_sb_info *sbi, s64 nblocks);
extern void ext4_free_blocks(handle_t *handle, struct inode *inode,
ext4_fsblk_t block, unsigned long count, int metadata);
-extern void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb,
- ext4_fsblk_t block, unsigned long count,
- unsigned long *pdquot_freed_blocks);
+extern void ext4_add_groupblocks(handle_t *handle, struct super_block *sb,
+ ext4_fsblk_t block, unsigned long count);
extern ext4_fsblk_t ext4_count_free_blocks(struct super_block *);
extern void ext4_check_blocks_bitmap(struct super_block *);
extern struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb,
diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c
index b6ec184..ecaacbb 100644
--- a/fs/ext4/resize.c
+++ b/fs/ext4/resize.c
@@ -975,9 +975,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
struct buffer_head *bh;
handle_t *handle;
int err;
- unsigned long freed_blocks;
ext4_group_t group;
- struct ext4_group_info *grp;
/* We don't need to worry about locking wrt other resizers just
* yet: we're going to revalidate es->s_blocks_count after
@@ -1076,7 +1074,8 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
unlock_super(sb);
ext4_debug("freeing blocks %llu through %llu\n", o_blocks_count,
o_blocks_count + add);
- ext4_free_blocks_sb(handle, sb, o_blocks_count, add, &freed_blocks);
+ /* We add the blocks to the bitmap and set the group need init bit */
+ ext4_add_groupblocks(handle, sb, o_blocks_count, add);
ext4_debug("freed blocks %llu through %llu\n", o_blocks_count,
o_blocks_count + add);
if ((err = ext4_journal_stop(handle)))
@@ -1119,12 +1118,6 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
ClearPageUptodate(page);
page_cache_release(page);
}
-
- /* Get the info on the last group */
- grp = ext4_get_group_info(sb, group);
The new groups added during resize are flagged as
need_init group. Make sure we properly initialize these
groups. When we have block size < page size and we are adding
new groups the page may still be marked uptodate even though
we haven't initialized the group.
Signed-off-by: Aneesh Kumar K.V <[email protected]>
---
fs/ext4/balloc.c | 21 +++--
fs/ext4/ext4.h | 2 +-
fs/ext4/mballoc.c | 218 ++++++++++++++++++++++++++++++++++++++---------------
fs/ext4/mballoc.h | 3 +
fs/ext4/resize.c | 41 +----------
5 files changed, 176 insertions(+), 109 deletions(-)
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index d598fd6..07266ec 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -381,6 +381,7 @@ void ext4_add_groupblocks(handle_t *handle, struct super_block *sb,
ext4_debug("Adding block(s) %llu-%llu\n", block, block + count - 1);
ext4_get_group_no_and_offset(sb, block, &block_group, &bit);
+ grp = ext4_get_group_info(sb, block_group);
/*
* Check to see if we are freeing blocks across a group
* boundary.
@@ -425,7 +426,11 @@ void ext4_add_groupblocks(handle_t *handle, struct super_block *sb,
err = ext4_journal_get_write_access(handle, gd_bh);
if (err)
goto error_return;
-
+ /*
+ * make sure we don't allow a parallel init on other groups in the
+ * same buddy cache
+ */
+ down_write(&grp->alloc_sem);
for (i = 0, blocks_freed = 0; i < count; i++) {
BUFFER_TRACE(bitmap_bh, "clear bit");
if (!ext4_clear_bit_atomic(sb_bgl_lock(sbi, block_group),
@@ -450,6 +455,13 @@ void ext4_add_groupblocks(handle_t *handle, struct super_block *sb,
sbi->s_flex_groups[flex_group].free_blocks += blocks_freed;
spin_unlock(sb_bgl_lock(sbi, flex_group));
}
+ /*
+ * request to reload the buddy with the
+ * new bitmap information
+ */
+ set_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &(grp->bb_state));
+ ext4_mb_update_group_info(grp, blocks_freed);
+ up_write(&grp->alloc_sem);
/* We dirtied the bitmap block */
BUFFER_TRACE(bitmap_bh, "dirtied bitmap block");
@@ -461,13 +473,6 @@ void ext4_add_groupblocks(handle_t *handle, struct super_block *sb,
if (!err)
err = ret;
sb->s_dirt = 1;
- /*
- * request to reload the buddy with the
- * new bitmap information
- */
- grp = ext4_get_group_info(sb, block_group);
- set_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &(grp->bb_state));
- ext4_mb_update_group_info(grp, blocks_freed);
error_return:
brelse(bitmap_bh);
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 004dd25..89c7426 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -1109,7 +1109,7 @@ extern int __init init_ext4_mballoc(void);
extern void exit_ext4_mballoc(void);
extern void ext4_mb_free_blocks(handle_t *, struct inode *,
unsigned long, unsigned long, int, unsigned long *);
-extern int ext4_mb_add_more_groupinfo(struct super_block *sb,
+extern int ext4_mb_add_groupinfo(struct super_block *sb,
ext4_group_t i, struct ext4_group_desc *desc);
extern void ext4_mb_update_group_info(struct ext4_group_info *grp,
ext4_grpblk_t add);
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 20c8b09..05631d1 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -886,18 +886,20 @@ static int ext4_mb_init_cache(struct page *page, char *incore)
ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
struct ext4_buddy *e4b)
{
- struct ext4_sb_info *sbi = EXT4_SB(sb);
- struct inode *inode = sbi->s_buddy_cache;
int blocks_per_page;
int block;
int pnum;
int poff;
struct page *page;
int ret;
+ struct ext4_group_info *grp;
+ struct ext4_sb_info *sbi = EXT4_SB(sb);
+ struct inode *inode = sbi->s_buddy_cache;
mb_debug("load group %lu\n", group);
blocks_per_page = PAGE_CACHE_SIZE / sb->s_blocksize;
+ grp = ext4_get_group_info(sb, group);
e4b->bd_blkbits = sb->s_blocksize_bits;
e4b->bd_info = ext4_get_group_info(sb, group);
@@ -905,6 +907,15 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
e4b->bd_group = group;
e4b->bd_buddy_page = NULL;
e4b->bd_bitmap_page = NULL;
+ e4b->alloc_semp = &grp->alloc_sem;
+
+ /* Take the read lock on the group alloc
+ * sem. This would make sure a parallel
+ * ext4_mb_init_group happening on other
+ * groups mapped by the page is blocked
+ * till we are done with allocation
+ */
+ down_read(e4b->alloc_semp);
/*
* the buddy cache inode stores the block bitmap
@@ -919,8 +930,15 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
* what we'd like to avoid in fast path ... */
page = find_get_page(inode->i_mapping, pnum);
if (page == NULL || !PageUptodate(page)) {
- if (page)
+ if (page) {
+ /* If the page is not uptodate
+ * that would imply nobody is using
+ * it. So we should be able to drop
+ * page_cache without being worried
+ * about other group allocation
+ */
page_cache_release(page);
+ }
page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS);
if (page) {
BUG_ON(page->mapping != inode->i_mapping);
@@ -985,6 +1003,9 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
page_cache_release(e4b->bd_buddy_page);
e4b->bd_buddy = NULL;
e4b->bd_bitmap = NULL;
+
+ /* Done with the buddy cache */
+ up_read(e4b->alloc_semp);
return ret;
}
@@ -994,6 +1015,8 @@ static void ext4_mb_release_desc(struct ext4_buddy *e4b)
page_cache_release(e4b->bd_bitmap_page);
if (e4b->bd_buddy_page)
page_cache_release(e4b->bd_buddy_page);
+ /* Done with the buddy cache */
+ up_read(e4b->alloc_semp);
}
@@ -1694,6 +1717,129 @@ static int ext4_mb_good_group(struct ext4_allocation_context *ac,
return 0;
}
+static int ext4_mb_init_group(struct super_block *sb, ext4_group_t group)
+{
+
+ int i, ret;
+ void *bitmap;
+ int blocks_per_page;
+ int groups_per_page;
+ int block, pnum, poff;
+ int num_grp_locked = 0;
+ ext4_group_t first_group;
+ struct ext4_group_info *grp, *this_grp;
+ struct ext4_sb_info *sbi = EXT4_SB(sb);
+ struct inode *inode = sbi->s_buddy_cache;
+ struct page *page = NULL, *bitmap_page = NULL;
+
+ mb_debug("init group %lu\n", group);
+ blocks_per_page = PAGE_CACHE_SIZE / sb->s_blocksize;
+ /*
+ * the buddy cache inode stores the block bitmap
+ * and buddy information in consecutive blocks.
+ * So for each group we need two blocks.
+ */
+ block = group * 2;
+ pnum = block / blocks_per_page;
+ poff = block % blocks_per_page;
+ first_group = pnum * blocks_per_page / 2;
+ this_grp = ext4_get_group_info(sb, group);
+
+ groups_per_page = blocks_per_page >> 1;
+ if (groups_per_page == 0)
+ groups_per_page = 1;
+
+ /* read all groups the page covers into the cache */
+ for (i = 0; i < groups_per_page; i++) {
+
+ if ((first_group + i) >= EXT4_SB(sb)->s_groups_count)
+ break;
+ grp = ext4_get_group_info(sb, first_group + i);
+ /* take all groups write allocation
+ * semaphore. This make sure there is
+ * no block allocation going on in any
+ * of that groups
+ */
+ down_write(&grp->alloc_sem);
+ }
+ /*
+ * make sure we look at only those groups
+ * that are locked. A resize can add more
+ * groups while this happen
+ */
+ num_grp_locked = i;
+ if (!EXT4_MB_GRP_NEED_INIT(this_grp)) {
+ /*
+ * somebody initialized the group
+ * return without doing anything
+ */
+ ret = 0;
+ goto err;
+ }
+
+ page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS);
+ if (page) {
+ BUG_ON(page->mapping != inode->i_mapping);
+ ret = ext4_mb_init_cache(page, NULL);
+ if (ret) {
+ unlock_page(page);
+ goto err;
+ }
+ unlock_page(page);
+ }
+ if (page == NULL || !PageUptodate(page)) {
+ ret = -EIO;
+ goto err;
+ }
+ mark_page_accessed(page);
+ bitmap_page = page;
+ bitmap = page_address(page) + (poff * sb->s_blocksize);
+
+ /* init buddy cache */
+ block++;
+ pnum = block / blocks_per_page;
+ poff = block % blocks_per_page;
+ page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS);
+ if (page == bitmap_page) {
+ /*
+ * If both the bitmap and buddy are in
+ * the same page we don't need to force
+ * init the buddy
+ */
+ unlock_page(page);
+ } else if (page) {
+ BUG_ON(page->mapping != inode->i_mapping);
+ ret = ext4_mb_init_cache(page, bitmap);
+ if (ret) {
+ unlock_page(page);
+ goto err;
+ }
+ unlock_page(page);
+ }
+ if (page == NULL || !PageUptodate(page)) {
+ ret = -EIO;
+ goto err;
+ }
+ mark_page_accessed(page);
+err:
+ /* release locks on all the groups */
+ for (i = 0; i < num_grp_locked; i++) {
+
+ grp = ext4_get_group_info(sb, first_group + i);
+ /* take all groups write allocation
+ * semaphore. This make sure there is
+ * no block allocation going on in any
+ * of that groups
+ */
+ up_write(&grp->alloc_sem);
+ }
+ if (bitmap_page)
+ page_cache_release(bitmap_page);
+ if (page)
+ page_cache_release(page);
+ return ret;
+}
+
static noinline_for_stack int
ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
{
@@ -1781,7 +1927,7 @@ ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
continue;
/* quick check to skip empty groups */
- grp = ext4_get_group_info(ac->ac_sb, group);
+ grp = ext4_get_group_info(sb, group);
if (grp->bb_free == 0)
continue;
@@ -1794,10 +1940,9 @@ ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
* we need full data about the group
* to make a good selection
*/
- err = ext4_mb_load_buddy(sb, group, &e4b);
+ err = ext4_mb_init_group(sb, group);
if (err)
goto out;
- ext4_mb_release_desc(&e4b);
}
/*
@@ -2248,7 +2393,7 @@ ext4_mb_store_history(struct ext4_allocation_context *ac)
/* Create and initialize ext4_group_info data for the given group. */
-static int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group,
+int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group,
struct ext4_group_desc *desc)
{
int i, len;
@@ -2306,6 +2451,7 @@ static int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group,
}
INIT_LIST_HEAD(&meta_group_info[i]->bb_prealloc_list);
+ init_rwsem(&meta_group_info[i]->alloc_sem);
meta_group_info[i]->bb_free_root.rb_node = NULL;;
#ifdef DOUBLE_CHECK
@@ -2333,54 +2479,6 @@ static int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group,
} /* ext4_mb_add_groupinfo */
/*
- * Add a group to the existing groups.
- * This function is used for online resize
- */
-int ext4_mb_add_more_groupinfo(struct super_block *sb, ext4_group_t group,
- struct ext4_group_desc *desc)
-{
- struct ext4_sb_info *sbi = EXT4_SB(sb);
- struct inode *inode = sbi->s_buddy_cache;
- int blocks_per_page;
- int block;
- int pnum;
- struct page *page;
- int err;
-
- /* Add group based on group descriptor*/
- err = ext4_mb_add_groupinfo(sb, group, desc);
- if (err)
- return err;
-
- /*
- * Cache pages containing dynamic mb_alloc datas (buddy and bitmap
- * datas) are set not up to date so that they will be re-initilaized
- * during the next call to ext4_mb_load_buddy
- */
-
- /* Set buddy page as not up to date */
- blocks_per_page = PAGE_CACHE_SIZE / sb->s_blocksize;
- block = group * 2;
- pnum = block / blocks_per_page;
- page = find_get_page(inode->i_mapping, pnum);
- if (page != NULL) {
- ClearPageUptodate(page);
- page_cache_release(page);
- }
-
- /* Set bitmap page as not up to date */
- block++;
- pnum = block / blocks_per_page;
- page = find_get_page(inode->i_mapping, pnum);
- if (page != NULL) {
- ClearPageUptodate(page);
- page_cache_release(page);
- }
-
- return 0;
-}
-
-/*
* Update an existing group.
* This function is used for online resize
*/
@@ -4588,11 +4686,6 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
err = ext4_journal_get_write_access(handle, gd_bh);
if (err)
goto error_return;
-
- err = ext4_mb_load_buddy(sb, block_group, &e4b);
- if (err)
- goto error_return;
-
#ifdef AGGRESSIVE_CHECK
{
int i;
@@ -4606,6 +4699,8 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
/* We dirtied the bitmap block */
BUFFER_TRACE(bitmap_bh, "dirtied bitmap block");
err = ext4_journal_dirty_metadata(handle, bitmap_bh);
+ if (err)
+ goto error_return;
if (ac) {
ac->ac_b_ex.fe_group = block_group;
@@ -4614,6 +4709,9 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
ext4_mb_store_history(ac);
}
+ err = ext4_mb_load_buddy(sb, block_group, &e4b);
+ if (err)
+ goto error_return;
if (metadata) {
/* blocks being freed are metadata. these blocks shouldn't
* be used until this transaction is committed */
diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h
index 557308a..7cc69c2 100644
--- a/fs/ext4/mballoc.h
+++ b/fs/ext4/mballoc.h
@@ -20,6 +20,7 @@
#include <linux/version.h>
#include <linux/blkdev.h>
#include <linux/marker.h>
+#include <linux/mutex.h>
#include "ext4_jbd2.h"
#include "ext4.h"
#include "group.h"
@@ -130,6 +131,7 @@ struct ext4_group_info {
#ifdef DOUBLE_CHECK
void *bb_bitmap;
#endif
+ struct rw_semaphore alloc_sem;
unsigned short bb_counters[];
};
@@ -251,6 +253,7 @@ struct ext4_buddy {
struct super_block *bd_sb;
__u16 bd_blkbits;
ext4_group_t bd_group;
+ struct rw_semaphore *alloc_semp;
};
#define EXT4_MB_BITMAP(e4b) ((e4b)->bd_bitmap)
#define EXT4_MB_BUDDY(e4b) ((e4b)->bd_buddy)
diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c
index ecaacbb..94ccf24 100644
--- a/fs/ext4/resize.c
+++ b/fs/ext4/resize.c
@@ -870,7 +870,7 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
* We can allocate memory for mb_alloc based on the new group
* descriptor
*/
- err = ext4_mb_add_more_groupinfo(sb, input->group, gdp);
+ err = ext4_mb_add_groupinfo(sb, input->group, gdp);
if (err)
goto exit_journal;
@@ -1081,45 +1081,6 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
if ((err = ext4_journal_stop(handle)))
goto exit_put;
- /*
- * Mark mballoc pages as not up to date so that they will be updated
- * next time they are loaded by ext4_mb_load_buddy.
- *
- * XXX Bad, Bad, BAD!!! We should not be overloading the
- * Uptodate flag, particularly on thte bitmap bh, as way of
- * hinting to ext4_mb_load_buddy() that it needs to be
- * overloaded. A user could take a LVM snapshot, then do an
- * on-line fsck, and clear the uptodate flag, and this would
- * not be a bug in userspace, but a bug in the kernel. FIXME!!!
- */
- {
- struct ext4_sb_info *sbi = EXT4_SB(sb);
- struct inode *inode = sbi->s_buddy_cache;
- int blocks_per_page;
- int block;
- int pnum;
- struct page *page;
-
- /* Set buddy page as not up to date */
- blocks_per_page = PAGE_CACHE_SIZE / sb->s_blocksize;
- block = group * 2;
- pnum = block / blocks_per_page;
- page = find_get_page(inode->i_mapping, pnum);
- if (page != NULL) {
- ClearPageUptodate(page);
- page_cache_release(page);
- }
-
- /* Set bitmap page as not up to date */
- block++;
- pnum = block / blocks_per_page;
- page = find_get_page(inode->i_mapping, pnum);
- if (page != NULL) {
- ClearPageUptodate(page);
- page_cache_release(page);
- }
- }
Move some of the forward declaration of the static functions
to mballoc.c where they are used. This enables us to include
mballoc.h in other .c files. Also correct the buddy cache
documentation.
Signed-off-by: Aneesh Kumar K.V <[email protected]>
---
fs/ext4/mballoc.c | 14 ++++++++++++--
fs/ext4/mballoc.h | 18 +-----------------
2 files changed, 13 insertions(+), 19 deletions(-)
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 05631d1..390dc81 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -100,7 +100,7 @@
* inode as:
*
* { page }
- * [ group 0 buddy][ group 0 bitmap] [group 1][ group 1]...
+ * [ group 0 bitmap][ group 0 buddy] [group 1][ group 1]...
*
*
* one block each for bitmap and buddy information. So for each group we
@@ -330,6 +330,16 @@
* object
*
*/
+static struct kmem_cache *ext4_pspace_cachep;
+static struct kmem_cache *ext4_ac_cachep;
+static struct kmem_cache *ext4_free_ext_cachep;
+static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap,
+ ext4_group_t group);
+static int ext4_mb_init_per_dev_proc(struct super_block *sb);
+static int ext4_mb_destroy_per_dev_proc(struct super_block *sb);
+static void release_blocks_on_commit(journal_t *journal, transaction_t *txn);
+
+
static inline void *mb_correct_addr_and_bit(int *bit, void *addr)
{
@@ -716,7 +726,7 @@ static void ext4_mb_generate_buddy(struct super_block *sb,
* stored in the inode as
*
* { page }
- * [ group 0 buddy][ group 0 bitmap] [group 1][ group 1]...
+ * [ group 0 bitmap][ group 0 buddy] [group 1][ group 1]...
*
*
* one block each for bitmap and buddy information.
diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h
index 7cc69c2..407b39a 100644
--- a/fs/ext4/mballoc.h
+++ b/fs/ext4/mballoc.h
@@ -99,9 +99,6 @@
*/
#define MB_DEFAULT_GROUP_PREALLOC 512
-static struct kmem_cache *ext4_pspace_cachep;
-static struct kmem_cache *ext4_ac_cachep;
-static struct kmem_cache *ext4_free_ext_cachep;
struct ext4_free_data {
/* this links the free block information from group_info */
@@ -263,25 +260,12 @@ static inline void ext4_mb_store_history(struct ext4_allocation_context *ac)
{
return;
}
-#else
-static void ext4_mb_store_history(struct ext4_allocation_context *ac);
#endif
#define in_range(b, first, len) ((b) >= (first) && (b) <= (first) + (len) - 1)
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,
- ext4_group_t group);
-static void ext4_mb_return_to_preallocation(struct inode *inode,
- struct ext4_buddy *e4b, sector_t block,
- int count);
-static void ext4_mb_put_pa(struct ext4_allocation_context *,
- struct super_block *, struct ext4_prealloc_space *pa);
-static int ext4_mb_init_per_dev_proc(struct super_block *sb);
-static int ext4_mb_destroy_per_dev_proc(struct super_block *sb);
-static void release_blocks_on_commit(journal_t *journal, transaction_t *txn);
sparse annotate the group info allocation semaphore
Signed-off-by: Aneesh Kumar K.V <[email protected]>
---
fs/ext4/mballoc.c | 28 +++++++++++++++++++++++-----
1 files changed, 23 insertions(+), 5 deletions(-)
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 390dc81..34a365e 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -895,6 +895,7 @@ static int ext4_mb_init_cache(struct page *page, char *incore)
static noinline_for_stack int
ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
struct ext4_buddy *e4b)
+__acquires(e4b->alloc_semp)
{
int blocks_per_page;
int block;
@@ -926,6 +927,7 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
* till we are done with allocation
*/
down_read(e4b->alloc_semp);
+ __acquire(e4b->alloc_semp);
/*
* the buddy cache inode stores the block bitmap
@@ -1020,6 +1022,7 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
}
static void ext4_mb_release_desc(struct ext4_buddy *e4b)
+__releases(e4b->alloc_semp)
{
if (e4b->bd_bitmap_page)
page_cache_release(e4b->bd_bitmap_page);
@@ -1027,6 +1030,7 @@ static void ext4_mb_release_desc(struct ext4_buddy *e4b)
page_cache_release(e4b->bd_buddy_page);
/* Done with the buddy cache */
up_read(e4b->alloc_semp);
+ __release(e4b->alloc_semp);
}
@@ -1468,8 +1472,10 @@ static int ext4_mb_try_best_found(struct ext4_allocation_context *ac,
BUG_ON(ex.fe_len <= 0);
err = ext4_mb_load_buddy(ac->ac_sb, group, e4b);
- if (err)
+ if (err) {
+ __release(e4b->alloc_semp);
return err;
+ }
ext4_lock_group(ac->ac_sb, group);
max = mb_find_extent(e4b, 0, ex.fe_start, ex.fe_len, &ex);
@@ -1499,8 +1505,10 @@ static int ext4_mb_find_by_goal(struct ext4_allocation_context *ac,
return 0;
err = ext4_mb_load_buddy(ac->ac_sb, group, e4b);
- if (err)
+ if (err) {
+ __release(e4b->alloc_semp);
return err;
+ }
ext4_lock_group(ac->ac_sb, group);
max = mb_find_extent(e4b, 0, ac->ac_g_ex.fe_start,
@@ -1963,8 +1971,10 @@ ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
continue;
err = ext4_mb_load_buddy(sb, group, &e4b);
- if (err)
+ if (err) {
+ __release(e4b->alloc_semp);
goto out;
+ }
ext4_lock_group(sb, group);
if (!ext4_mb_good_group(ac, group, cr)) {
@@ -2275,6 +2285,7 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v)
sizeof(struct ext4_group_info);
err = ext4_mb_load_buddy(sb, group, &e4b);
if (err) {
+ __release(e4b->alloc_semp);
seq_printf(seq, "#%-5lu: I/O error\n", group);
return 0;
}
@@ -3820,6 +3831,7 @@ ext4_mb_discard_group_preallocations(struct super_block *sb,
err = ext4_mb_load_buddy(sb, group, &e4b);
if (err) {
+ __release(e4b->alloc_semp);
ext4_error(sb, __func__, "Error in loading buddy "
"information for %lu\n", group);
put_bh(bitmap_bh);
@@ -3987,6 +3999,7 @@ void ext4_discard_preallocations(struct inode *inode)
err = ext4_mb_load_buddy(sb, group, &e4b);
if (err) {
+ __release(e4b->alloc_semp);
ext4_error(sb, __func__, "Error in loading buddy "
"information for %lu\n", group);
continue;
@@ -4260,6 +4273,7 @@ ext4_mb_discard_lg_preallocations(struct super_block *sb,
ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, NULL);
if (ext4_mb_load_buddy(sb, group, &e4b)) {
+ __release(e4b->alloc_semp);
ext4_error(sb, __func__, "Error in loading buddy "
"information for %lu\n", group);
continue;
@@ -4720,8 +4734,10 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
}
err = ext4_mb_load_buddy(sb, block_group, &e4b);
- if (err)
+ if (err) {
+ __release(e4b->alloc_semp);
goto error_return;
+ }
if (metadata) {
/* blocks being freed are metadata. these blocks shouldn't
* be used until this transaction is committed */
@@ -4805,8 +4821,10 @@ ext4_mb_search_free_extents(struct inode *inode,
end = EXT4_BLOCKS_PER_GROUP(sb) - 1;
err = ext4_mb_load_buddy(sb, group_no, &e4b);
- if (err)
+ if (err) {
+ __release(e4b->alloc_semp);
return err;
+ }
/* We consider about the boot block if bs = 1k */
if (sb->s_blocksize == 1024)
--
1.6.0.3.514.g2f91b
Signed-off-by: Aneesh Kumar K.V <[email protected]>
---
fs/jbd2/checkpoint.c | 6 ++++--
fs/jbd2/commit.c | 16 ++++++++++------
include/linux/jbd2.h | 9 +++++++--
3 files changed, 21 insertions(+), 10 deletions(-)
diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index 9203c33..cbe7227 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -716,7 +716,8 @@ void __jbd2_journal_insert_checkpoint(struct journal_head *jh,
* Called with j_list_lock held.
*/
-void __jbd2_journal_drop_transaction(journal_t *journal, transaction_t *transaction)
+void ___jbd2_journal_drop_transaction(journal_t *journal,
+ transaction_t *transaction, int free)
{
assert_spin_locked(&journal->j_list_lock);
if (transaction->t_cpnext) {
@@ -742,5 +743,6 @@ void __jbd2_journal_drop_transaction(journal_t *journal, transaction_t *transact
J_ASSERT(journal->j_running_transaction != transaction);
jbd_debug(1, "Dropping transaction %d, all done\n", transaction->t_tid);
- kfree(transaction);
+ if (free)
+ kfree(transaction);
}
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index ebc667b..35f016c 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -338,7 +338,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
int space_left = 0;
int first_tag = 0;
int tag_flag;
- int i;
+ int i, to_free = 0;
int tag_bytes = journal_tag_bytes(journal);
struct buffer_head *cbh = NULL; /* For transactional checksums */
__u32 crc32_sum = ~0;
@@ -974,12 +974,11 @@ void jbd2_journal_commit_transaction(journal_t *journal)
journal->j_committing_transaction = NULL;
spin_unlock(&journal->j_state_lock);
- if (journal->j_commit_callback)
- journal->j_commit_callback(journal, commit_transaction);
-
if (commit_transaction->t_checkpoint_list == NULL &&
commit_transaction->t_checkpoint_io_list == NULL) {
- __jbd2_journal_drop_transaction(journal, commit_transaction);
+ ___jbd2_journal_drop_transaction(journal,
+ commit_transaction, 0);
+ to_free = 1;
} else {
if (journal->j_checkpoint_transactions == NULL) {
journal->j_checkpoint_transactions = commit_transaction;
@@ -998,11 +997,16 @@ void jbd2_journal_commit_transaction(journal_t *journal)
}
spin_unlock(&journal->j_list_lock);
+ if (journal->j_commit_callback)
+ journal->j_commit_callback(journal, commit_transaction);
+
trace_mark(jbd2_end_commit, "dev %s transaction %d head %d",
- journal->j_devname, journal->j_commit_sequence,
+ journal->j_devname, commit_transaction->t_tid,
journal->j_tail_sequence);
jbd_debug(1, "JBD: commit %d complete, head %d\n",
journal->j_commit_sequence, journal->j_tail_sequence);
+ if (to_free)
+ kfree(commit_transaction);
wake_up(&journal->j_wait_done_commit);
}
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index 973db89..0ff7209 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -1146,8 +1146,8 @@ int jbd2_log_wait_commit(journal_t *journal, tid_t tid);
int jbd2_log_do_checkpoint(journal_t *journal);
void __jbd2_log_wait_for_space(journal_t *journal);
-extern void __jbd2_journal_drop_transaction(journal_t *, transaction_t *);
-extern int jbd2_cleanup_journal_tail(journal_t *);
+extern void ___jbd2_journal_drop_transaction(journal_t *, transaction_t *, int);
+extern int jbd2_cleanup_journal_tail(journal_t *);
/* Debugging code only: */
@@ -1217,6 +1217,11 @@ static inline int jbd_space_needed(journal_t *journal)
t_outstanding_credits;
return nblocks;
}
+static inline void __jbd2_journal_drop_transaction(journal_t *journal,
+ transaction_t *transaction)
+{
+ return ___jbd2_journal_drop_transaction(journal, transaction, 1);
+}
/*
* Definitions which augment the buffer_head layer
--
1.6.0.3.514.g2f91b
When we generate buddy cache(especially during resize) we need to make
sure we don't use the blocks freed but not yet comitted. This make
sure we have the right value of free blocks count in the group
info and also in the bitmap. This also ensures the ordered mode
consistency
Signed-off-by: Aneesh Kumar K.V <[email protected]>
---
fs/ext4/mballoc.c | 69 ++++++++++++++++++++++++++++++++++++++++++++--------
1 files changed, 58 insertions(+), 11 deletions(-)
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 34a365e..5824170 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -335,6 +335,9 @@
static struct kmem_cache *ext4_free_ext_cachep;
static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap,
ext4_group_t group);
+static void ext4_mb_generate_from_freelist(struct super_block *sb, void *bitmap,
+ ext4_group_t group,
+ struct ext4_free_data *entry);
static int ext4_mb_init_per_dev_proc(struct super_block *sb);
static int ext4_mb_destroy_per_dev_proc(struct super_block *sb);
static void release_blocks_on_commit(journal_t *journal, transaction_t *txn);
@@ -858,7 +861,9 @@ static int ext4_mb_init_cache(struct page *page, char *incore)
/*
* incore got set to the group block bitmap below
*/
+ ext4_lock_group(sb, group);
ext4_mb_generate_buddy(sb, data, incore, group);
+ ext4_unlock_group(sb, group);
incore = NULL;
} else {
/* this is block of bitmap */
@@ -872,6 +877,7 @@ static int ext4_mb_init_cache(struct page *page, char *incore)
/* mark all preallocated blks used in in-core bitmap */
ext4_mb_generate_from_pa(sb, data, group);
+ ext4_mb_generate_from_freelist(sb, data, group, NULL);
ext4_unlock_group(sb, group);
/* set incore so that the buddy information can be
@@ -3432,6 +3438,42 @@ ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
}
/*
+ * the function goes through all block freed in the group
+ * but not yet committed 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_freelist(struct super_block *sb, void *bitmap,
+ ext4_group_t group,
+ struct ext4_free_data *entry)
+{
+ struct rb_node *n;
+ struct ext4_group_info *grp;
+ struct ext4_free_data *new_entry;
+ if (entry == NULL) {
+ grp = ext4_get_group_info(sb, group);
+ n = grp->bb_free_root.rb_node;
+ } else
+ n = &entry->node;
+
+ if (n == NULL)
+ return;
+ if (n->rb_left) {
+ new_entry = rb_entry(n->rb_left, struct ext4_free_data, node);
+ ext4_mb_generate_from_freelist(sb, bitmap, group, new_entry);
+ }
+ if (n->rb_right) {
+ new_entry = rb_entry(n->rb_right, struct ext4_free_data, node);
+ ext4_mb_generate_from_freelist(sb, bitmap, group, new_entry);
+ }
+ printk(KERN_CRIT "Aneesh updating bitmap");
+ mb_set_bits(sb_bgl_lock(EXT4_SB(sb), group),
+ bitmap, entry->start_blk,
+ entry->count);
+ return;
+}
+
+/*
* 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)
@@ -4717,15 +4759,6 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
BUG_ON(!mb_test_bit(bit + i, bitmap_bh->b_data));
}
#endif
- mb_clear_bits(sb_bgl_lock(sbi, block_group), bitmap_bh->b_data,
- bit, count);
-
- /* We dirtied the bitmap block */
- BUFFER_TRACE(bitmap_bh, "dirtied bitmap block");
- err = ext4_journal_dirty_metadata(handle, bitmap_bh);
- if (err)
- goto error_return;
-
if (ac) {
ac->ac_b_ex.fe_group = block_group;
ac->ac_b_ex.fe_start = bit;
@@ -4739,11 +4772,21 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
goto error_return;
}
if (metadata) {
- /* blocks being freed are metadata. these blocks shouldn't
- * be used until this transaction is committed */
+ /*
+ * blocks being freed are metadata. these blocks shouldn't
+ * be used until this transaction is committed
+ */
+ mb_clear_bits(sb_bgl_lock(sbi, block_group), bitmap_bh->b_data,
+ bit, count);
ext4_mb_free_metadata(handle, &e4b, block_group, bit, count);
} else {
ext4_lock_group(sb, block_group);
+ /* need to update group_info->bb_free and bitmap
+ * with group lock held. generate_buddy look at
+ * them with group lock_held
+ */
+ mb_clear_bits(sb_bgl_lock(sbi, block_group), bitmap_bh->b_data,
+ bit, count);
mb_free_blocks(inode, &e4b, bit, count);
ext4_mb_return_to_preallocation(inode, &e4b, block, count);
ext4_unlock_group(sb, block_group);
@@ -4766,6 +4809,10 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
*freed += count;
+ /* We dirtied the bitmap block */
+ BUFFER_TRACE(bitmap_bh, "dirtied bitmap block");
+ err = ext4_journal_dirty_metadata(handle, bitmap_bh);
+
/* And the group descriptor block */
BUFFER_TRACE(gd_bh, "dirtied group descriptor block");
ret = ext4_journal_dirty_metadata(handle, gd_bh);
--
1.6.0.3.514.g2f91b
On Fri, Oct 31, 2008 at 10:20:54PM +0530, Aneesh Kumar K.V wrote:
> With this change new blocks added during resize
> are marked as free in the block bitmap and the
> group is flagged with EXT4_GROUP_INFO_NEED_INIT_BIT
> flag. This make sure when mballoc tries to allocate
> blocks from the new group we would reload the
> buddy information using the bitmap present in the disk.
My comment about why not use ext4_mb_free_blocks() still applies.
Also, if you're going to replace a patch, please put V2 (or Vn+1) in
the subject like. I wasted time dealing with your earlier set of
three resize patches since I didn't realize they were replicated in
this six patch series.
Also, Patch 2 of 6 didn't apply as of 2.6.28-rc3. I fixed it up in
the patch queue, but please check it.
Thanks!!
- Ted
Justification for this patch, please? Is there a lockdep violation
that you're trying to avoid?
Also, given that __jbd2_journal_drop_transaction() is called exactly
twice, instead of passing in an extra free, it might be simpler just
to transfer the responsibility of calling kfree() to the caller.
- Ted
On Fri, Oct 31, 2008 at 10:20:59PM +0530, Aneesh Kumar K.V wrote:
> When we generate buddy cache(especially during resize) we need to make
> sure we don't use the blocks freed but not yet comitted. This make
> sure we have the right value of free blocks count in the group
> info and also in the bitmap. This also ensures the ordered mode
> consistency
Wouldn't this be fixed if we have resize call ext4_mb_free_blocks()
instead of having it regeneerate the buddy cache?
- Ted