* mballoc.c contains the multiblocks allocation routines
*/
-#include <linux/time.h>
-#include <linux/fs.h>
-#include <linux/namei.h>
-#include <linux/quotaops.h>
-#include <linux/buffer_head.h>
-#include <linux/module.h>
-#include <linux/swap.h>
-#include <linux/proc_fs.h>
-#include <linux/pagemap.h>
-#include <linux/seq_file.h>
-#include <linux/version.h>
-#include "ext4_jbd2.h"
-#include "ext4.h"
-#include "group.h"
-
+#include "mballoc.h"
/*
* MUSTDO:
* - test ext4_ext_search_left() and ext4_ext_search_right()
*
*/
-/*
- * with AGGRESSIVE_CHECK allocator runs consistency checks over
- * structures. these checks slow things down a lot
- */
-#define AGGRESSIVE_CHECK__
-
-/*
- * with DOUBLE_CHECK defined mballoc creates persistent in-core
- * bitmaps, maintains and uses them to check for double allocations
- */
-#define DOUBLE_CHECK__
-
-/*
- */
-#define MB_DEBUG__
-#ifdef MB_DEBUG
-#define mb_debug(fmt, a...) printk(fmt, ##a)
-#else
-#define mb_debug(fmt, a...)
-#endif
-
-/*
- * with EXT4_MB_HISTORY mballoc stores last N allocations in memory
- * and you can monitor it in /proc/fs/ext4/<dev>/mb_history
- */
-#define EXT4_MB_HISTORY
-#define EXT4_MB_HISTORY_ALLOC 1 /* allocation */
-#define EXT4_MB_HISTORY_PREALLOC 2 /* preallocated blocks used */
-#define EXT4_MB_HISTORY_DISCARD 4 /* preallocation discarded */
-#define EXT4_MB_HISTORY_FREE 8 /* free */
-
-#define EXT4_MB_HISTORY_DEFAULT (EXT4_MB_HISTORY_ALLOC | \
- EXT4_MB_HISTORY_PREALLOC)
-
-/*
- * How long mballoc can look for a best extent (in found extents)
- */
-#define MB_DEFAULT_MAX_TO_SCAN 200
-
-/*
- * How long mballoc must look for a best extent
- */
-#define MB_DEFAULT_MIN_TO_SCAN 10
-
-/*
- * How many groups mballoc will scan looking for the best chunk
- */
-#define MB_DEFAULT_MAX_GROUPS_TO_SCAN 5
-
-/*
- * with 'ext4_mb_stats' allocator will collect stats that will be
- * shown at umount. The collecting costs though!
- */
-#define MB_DEFAULT_STATS 1
-
-/*
- * files smaller than MB_DEFAULT_STREAM_THRESHOLD are served
- * by the stream allocator, which purpose is to pack requests
- * as close each to other as possible to produce smooth I/O traffic
- * We use locality group prealloc space for stream request.
- * We can tune the same via /proc/fs/ext4/<parition>/stream_req
- */
-#define MB_DEFAULT_STREAM_THRESHOLD 16 /* 64K */
-
-/*
- * for which requests use 2^N search using buddies
- */
-#define MB_DEFAULT_ORDER2_REQS 2
-
-/*
- * default group prealloc size 512 blocks
- */
-#define MB_DEFAULT_GROUP_PREALLOC 512
-
-static struct kmem_cache *ext4_pspace_cachep;
-static struct kmem_cache *ext4_ac_cachep;
-
-#ifdef EXT4_BB_MAX_BLOCKS
-#undef EXT4_BB_MAX_BLOCKS
-#endif
-#define EXT4_BB_MAX_BLOCKS 30
-
-struct ext4_free_metadata {
- ext4_group_t group;
- unsigned short num;
- ext4_grpblk_t blocks[EXT4_BB_MAX_BLOCKS];
- struct list_head list;
-};
-
-struct ext4_group_info {
- unsigned long bb_state;
- unsigned long bb_tid;
- struct ext4_free_metadata *bb_md_cur;
- unsigned short bb_first_free;
- unsigned short bb_free;
- unsigned short bb_fragments;
- struct list_head bb_prealloc_list;
-#ifdef DOUBLE_CHECK
- void *bb_bitmap;
-#endif
- unsigned short bb_counters[];
-};
-
-#define EXT4_GROUP_INFO_NEED_INIT_BIT 0
-#define EXT4_GROUP_INFO_LOCKED_BIT 1
-
-#define EXT4_MB_GRP_NEED_INIT(grp) \
- (test_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &((grp)->bb_state)))
-
-
-struct ext4_prealloc_space {
- struct list_head pa_inode_list;
- struct list_head pa_group_list;
- union {
- struct list_head pa_tmp_list;
- struct rcu_head pa_rcu;
- } u;
- spinlock_t pa_lock;
- atomic_t pa_count;
- unsigned pa_deleted;
- ext4_fsblk_t pa_pstart; /* phys. block */
- ext4_lblk_t pa_lstart; /* log. block */
- unsigned short pa_len; /* len of preallocated chunk */
- unsigned short pa_free; /* how many blocks are free */
- unsigned short pa_linear; /* consumed in one direction
- * strictly, for grp prealloc */
- spinlock_t *pa_obj_lock;
- struct inode *pa_inode; /* hack, for history only */
-};
-
-
-struct ext4_free_extent {
- ext4_lblk_t fe_logical;
- ext4_grpblk_t fe_start;
- ext4_group_t fe_group;
- int fe_len;
-};
-
-/*
- * Locality group:
- * we try to group all related changes together
- * so that writeback can flush/allocate them together as well
- */
-struct ext4_locality_group {
- /* for allocator */
- struct mutex lg_mutex; /* to serialize allocates */
- struct list_head lg_prealloc_list;/* list of preallocations */
- spinlock_t lg_prealloc_lock;
-};
-
-struct ext4_allocation_context {
- struct inode *ac_inode;
- struct super_block *ac_sb;
-
- /* original request */
- struct ext4_free_extent ac_o_ex;
-
- /* goal request (after normalization) */
- struct ext4_free_extent ac_g_ex;
-
- /* the best found extent */
- struct ext4_free_extent ac_b_ex;
-
- /* copy of the bext found extent taken before preallocation efforts */
- struct ext4_free_extent ac_f_ex;
-
- /* number of iterations done. we have to track to limit searching */
- unsigned long ac_ex_scanned;
- __u16 ac_groups_scanned;
- __u16 ac_found;
- __u16 ac_tail;
- __u16 ac_buddy;
- __u16 ac_flags; /* allocation hints */
- __u8 ac_status;
- __u8 ac_criteria;
- __u8 ac_repeats;
- __u8 ac_2order; /* if request is to allocate 2^N blocks and
- * N > 0, the field stores N, otherwise 0 */
- __u8 ac_op; /* operation, for history only */
- struct page *ac_bitmap_page;
- struct page *ac_buddy_page;
- struct ext4_prealloc_space *ac_pa;
- struct ext4_locality_group *ac_lg;
-};
-
-#define AC_STATUS_CONTINUE 1
-#define AC_STATUS_FOUND 2
-#define AC_STATUS_BREAK 3
-
-struct ext4_mb_history {
- struct ext4_free_extent orig; /* orig allocation */
- struct ext4_free_extent goal; /* goal allocation */
- struct ext4_free_extent result; /* result allocation */
- unsigned pid;
- unsigned ino;
- __u16 found; /* how many extents have been found */
- __u16 groups; /* how many groups have been scanned */
- __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 op:4;
- __u8 merged:1;
-};
-
-struct ext4_buddy {
- struct page *bd_buddy_page;
- void *bd_buddy;
- struct page *bd_bitmap_page;
- void *bd_bitmap;
- struct ext4_group_info *bd_info;
- struct super_block *bd_sb;
- __u16 bd_blkbits;
- ext4_group_t bd_group;
-};
-#define EXT4_MB_BITMAP(e4b) ((e4b)->bd_bitmap)
-#define EXT4_MB_BUDDY(e4b) ((e4b)->bd_buddy)
-
-#ifndef EXT4_MB_HISTORY
-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)
-
-static struct proc_dir_entry *proc_root_ext4;
-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_poll_new_transaction(struct super_block *, handle_t *);
-static void ext4_mb_free_committed_blocks(struct super_block *);
-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 inline void ext4_lock_group(struct super_block *sb, ext4_group_t group)
-{
- struct ext4_group_info *grinfo = ext4_get_group_info(sb, group);
-
- bit_spin_lock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state));
-}
-
-static inline void ext4_unlock_group(struct super_block *sb,
- ext4_group_t group)
-{
- struct ext4_group_info *grinfo = ext4_get_group_info(sb, group);
-
- bit_spin_unlock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state));
-}
-
-static inline int ext4_is_group_locked(struct super_block *sb,
- ext4_group_t group)
-{
- struct ext4_group_info *grinfo = ext4_get_group_info(sb, group);
-
- return bit_spin_is_locked(EXT4_GROUP_INFO_LOCKED_BIT,
- &(grinfo->bb_state));
-}
-
-static ext4_fsblk_t ext4_grp_offs_to_block(struct super_block *sb,
- struct ext4_free_extent *fex)
-{
- ext4_fsblk_t block;
-
- block = (ext4_fsblk_t) fex->fe_group * EXT4_BLOCKS_PER_GROUP(sb)
- + fex->fe_start
- + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
- return block;
-}
-
static inline void *mb_correct_addr_and_bit(int *bit, void *addr)
{
#if BITS_PER_LONG == 64
static inline int mb_find_next_zero_bit(void *addr, int max, int start)
{
- int fix = 0;
+ int fix = 0, ret, tmpmax;
addr = mb_correct_addr_and_bit(&fix, addr);
- max += fix;
+ tmpmax = max + fix;
start += fix;
- return ext4_find_next_zero_bit(addr, max, start) - fix;
+ ret = ext4_find_next_zero_bit(addr, tmpmax, start) - fix;
+ if (ret > max)
+ return max;
+ return ret;
}
static inline int mb_find_next_bit(void *addr, int max, int start)
{
- int fix = 0;
+ int fix = 0, ret, tmpmax;
addr = mb_correct_addr_and_bit(&fix, addr);
- max += fix;
+ tmpmax = max + fix;
start += fix;
- return ext4_find_next_bit(addr, max, start) - fix;
+ ret = ext4_find_next_bit(addr, tmpmax, start) - fix;
+ if (ret > max)
+ return max;
+ return ret;
}
static void *mb_find_buddy(struct ext4_buddy *e4b, int order, int *max)
if (!buffer_uptodate(bh[i]))
goto out;
+ err = 0;
first_block = page->index * blocks_per_page;
for (i = 0; i < blocks_per_page; i++) {
int group;
int pnum;
int poff;
struct page *page;
+ int ret;
mb_debug("load group %lu\n", group);
if (page) {
BUG_ON(page->mapping != inode->i_mapping);
if (!PageUptodate(page)) {
- ext4_mb_init_cache(page, NULL);
+ ret = ext4_mb_init_cache(page, NULL);
+ if (ret) {
+ unlock_page(page);
+ goto err;
+ }
mb_cmp_bitmaps(e4b, page_address(page) +
(poff * sb->s_blocksize));
}
unlock_page(page);
}
}
- if (page == NULL || !PageUptodate(page))
+ if (page == NULL || !PageUptodate(page)) {
+ ret = -EIO;
goto err;
+ }
e4b->bd_bitmap_page = page;
e4b->bd_bitmap = page_address(page) + (poff * sb->s_blocksize);
mark_page_accessed(page);
page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS);
if (page) {
BUG_ON(page->mapping != inode->i_mapping);
- if (!PageUptodate(page))
- ext4_mb_init_cache(page, e4b->bd_bitmap);
-
+ if (!PageUptodate(page)) {
+ ret = ext4_mb_init_cache(page, e4b->bd_bitmap);
+ if (ret) {
+ unlock_page(page);
+ goto err;
+ }
+ }
unlock_page(page);
}
}
- if (page == NULL || !PageUptodate(page))
+ if (page == NULL || !PageUptodate(page)) {
+ ret = -EIO;
goto err;
+ }
e4b->bd_buddy_page = page;
e4b->bd_buddy = page_address(page) + (poff * sb->s_blocksize);
mark_page_accessed(page);
page_cache_release(e4b->bd_buddy_page);
e4b->bd_buddy = NULL;
e4b->bd_bitmap = NULL;
- return -EIO;
+ return ret;
}
static void ext4_mb_release_desc(struct ext4_buddy *e4b)
}
}
-static int mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b,
+static void mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b,
int first, int count)
{
int block = 0;
blocknr += block;
blocknr +=
le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
-
+ ext4_unlock_group(sb, e4b->bd_group);
ext4_error(sb, __func__, "double-free of inode"
" %lu's block %llu(bit %u in group %lu)\n",
inode ? inode->i_ino : 0, blocknr, block,
e4b->bd_group);
+ ext4_lock_group(sb, e4b->bd_group);
}
mb_clear_bit(block, EXT4_MB_BITMAP(e4b));
e4b->bd_info->bb_counters[order]++;
} while (1);
}
mb_check_buddy(e4b);
-
- return 0;
}
static int mb_find_extent(struct ext4_buddy *e4b, int order, int block,
ac->ac_g_ex.fe_start = sbi->s_mb_last_start;
spin_unlock(&sbi->s_md_lock);
}
-
- /* searching for the right group start from the goal value specified */
- group = ac->ac_g_ex.fe_group;
-
/* Let's just scan groups to find more-less suitable blocks */
cr = ac->ac_2order ? 0 : 1;
/*
repeat:
for (; cr < 4 && ac->ac_status == AC_STATUS_CONTINUE; cr++) {
ac->ac_criteria = cr;
+ /*
+ * searching for the right group start
+ * from the goal value specified
+ */
+ group = ac->ac_g_ex.fe_group;
+
for (i = 0; i < EXT4_SB(sb)->s_groups_count; group++, i++) {
struct ext4_group_info *grp;
struct ext4_group_desc *desc;
int rc;
int size;
+ if (unlikely(sbi->s_mb_history == NULL))
+ return -ENOMEM;
s = kmalloc(sizeof(*s), GFP_KERNEL);
if (s == NULL)
return -ENOMEM;
sbi->s_mb_history_cur = 0;
spin_lock_init(&sbi->s_mb_history_lock);
i = sbi->s_mb_history_max * sizeof(struct ext4_mb_history);
- sbi->s_mb_history = kmalloc(i, GFP_KERNEL);
- if (likely(sbi->s_mb_history != NULL))
- memset(sbi->s_mb_history, 0, i);
+ sbi->s_mb_history = kzalloc(i, GFP_KERNEL);
/* if we can't allocate history, then we simple won't use it */
}
#define ext4_mb_history_init(sb)
#endif
+
+/* Create and initialize ext4_group_info data for the given group. */
+int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group,
+ struct ext4_group_desc *desc)
+{
+ int i, len;
+ int metalen = 0;
+ struct ext4_sb_info *sbi = EXT4_SB(sb);
+ struct ext4_group_info **meta_group_info;
+
+ /*
+ * First check if this group is the first of a reserved block.
+ * If it's true, we have to allocate a new table of pointers
+ * to ext4_group_info structures
+ */
+ if (group % EXT4_DESC_PER_BLOCK(sb) == 0) {
+ metalen = sizeof(*meta_group_info) <<
+ EXT4_DESC_PER_BLOCK_BITS(sb);
+ meta_group_info = kmalloc(metalen, GFP_KERNEL);
+ if (meta_group_info == NULL) {
+ printk(KERN_ERR "EXT4-fs: can't allocate mem for a "
+ "buddy group\n");
+ goto exit_meta_group_info;
+ }
+ sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)] =
+ meta_group_info;
+ }
+
+ /*
+ * calculate needed size. if change bb_counters size,
+ * don't forget about ext4_mb_generate_buddy()
+ */
+ len = offsetof(typeof(**meta_group_info),
+ bb_counters[sb->s_blocksize_bits + 2]);
+
+ meta_group_info =
+ sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)];
+ i = group & (EXT4_DESC_PER_BLOCK(sb) - 1);
+
+ meta_group_info[i] = kzalloc(len, GFP_KERNEL);
+ if (meta_group_info[i] == NULL) {
+ printk(KERN_ERR "EXT4-fs: can't allocate buddy mem\n");
+ goto exit_group_info;
+ }
+ set_bit(EXT4_GROUP_INFO_NEED_INIT_BIT,
+ &(meta_group_info[i]->bb_state));
+
+ /*
+ * initialize bb_free to be able to skip
+ * empty groups without initialization
+ */
+ if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
+ meta_group_info[i]->bb_free =
+ ext4_free_blocks_after_init(sb, group, desc);
+ } else {
+ meta_group_info[i]->bb_free =
+ le16_to_cpu(desc->bg_free_blocks_count);
+ }
+
+ INIT_LIST_HEAD(&meta_group_info[i]->bb_prealloc_list);
+
+#ifdef DOUBLE_CHECK
+ {
+ struct buffer_head *bh;
+ meta_group_info[i]->bb_bitmap =
+ kmalloc(sb->s_blocksize, GFP_KERNEL);
+ BUG_ON(meta_group_info[i]->bb_bitmap == NULL);
+ bh = ext4_read_block_bitmap(sb, group);
+ BUG_ON(bh == NULL);
+ memcpy(meta_group_info[i]->bb_bitmap, bh->b_data,
+ sb->s_blocksize);
+ put_bh(bh);
+ }
+#endif
+
+ return 0;
+
+exit_group_info:
+ /* If a meta_group_info table has been allocated, release it now */
+ if (group % EXT4_DESC_PER_BLOCK(sb) == 0)
+ kfree(sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)]);
+exit_meta_group_info:
+ return -ENOMEM;
+} /* 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
+ */
+void ext4_mb_update_group_info(struct ext4_group_info *grp, ext4_grpblk_t add)
+{
+ grp->bb_free += add;
+}
+
static int ext4_mb_init_backend(struct super_block *sb)
{
ext4_group_t i;
- int j, len, metalen;
+ int metalen;
struct ext4_sb_info *sbi = EXT4_SB(sb);
- int num_meta_group_infos =
- (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) >>
- EXT4_DESC_PER_BLOCK_BITS(sb);
+ struct ext4_super_block *es = sbi->s_es;
+ int num_meta_group_infos;
+ int num_meta_group_infos_max;
+ int array_size;
struct ext4_group_info **meta_group_info;
+ struct ext4_group_desc *desc;
+ /* This is the number of blocks used by GDT */
+ num_meta_group_infos = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) -
+ 1) >> EXT4_DESC_PER_BLOCK_BITS(sb);
+
+ /*
+ * This is the total number of blocks used by GDT including
+ * the number of reserved blocks for GDT.
+ * The s_group_info array is allocated with this value
+ * to allow a clean online resize without a complex
+ * manipulation of pointer.
+ * The drawback is the unused memory when no resize
+ * occurs but it's very low in terms of pages
+ * (see comments below)
+ * Need to handle this properly when META_BG resizing is allowed
+ */
+ num_meta_group_infos_max = num_meta_group_infos +
+ le16_to_cpu(es->s_reserved_gdt_blocks);
+
+ /*
+ * array_size is the size of s_group_info array. We round it
+ * to the next power of two because this approximation is done
+ * internally by kmalloc so we can have some more memory
+ * for free here (e.g. may be used for META_BG resize).
+ */
+ array_size = 1;
+ while (array_size < sizeof(*sbi->s_group_info) *
+ num_meta_group_infos_max)
+ array_size = array_size << 1;
/* An 8TB filesystem with 64-bit pointers requires a 4096 byte
* kmalloc. A 128kb malloc should suffice for a 256TB filesystem.
* So a two level scheme suffices for now. */
- sbi->s_group_info = kmalloc(sizeof(*sbi->s_group_info) *
- num_meta_group_infos, GFP_KERNEL);
+ sbi->s_group_info = kmalloc(array_size, GFP_KERNEL);
if (sbi->s_group_info == NULL) {
printk(KERN_ERR "EXT4-fs: can't allocate buddy meta group\n");
return -ENOMEM;
sbi->s_group_info[i] = meta_group_info;
}
- /*
- * calculate needed size. if change bb_counters size,
- * don't forget about ext4_mb_generate_buddy()
- */
- len = sizeof(struct ext4_group_info);
- len += sizeof(unsigned short) * (sb->s_blocksize_bits + 2);
for (i = 0; i < sbi->s_groups_count; i++) {
- struct ext4_group_desc *desc;
-
- meta_group_info =
- sbi->s_group_info[i >> EXT4_DESC_PER_BLOCK_BITS(sb)];
- j = i & (EXT4_DESC_PER_BLOCK(sb) - 1);
-
- meta_group_info[j] = kzalloc(len, GFP_KERNEL);
- if (meta_group_info[j] == NULL) {
- printk(KERN_ERR "EXT4-fs: can't allocate buddy mem\n");
- i--;
- goto err_freebuddy;
- }
desc = ext4_get_group_desc(sb, i, NULL);
if (desc == NULL) {
printk(KERN_ERR
"EXT4-fs: can't read descriptor %lu\n", i);
goto err_freebuddy;
}
- memset(meta_group_info[j], 0, len);
- set_bit(EXT4_GROUP_INFO_NEED_INIT_BIT,
- &(meta_group_info[j]->bb_state));
-
- /*
- * initialize bb_free to be able to skip
- * empty groups without initialization
- */
- if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
- meta_group_info[j]->bb_free =
- ext4_free_blocks_after_init(sb, i, desc);
- } else {
- meta_group_info[j]->bb_free =
- le16_to_cpu(desc->bg_free_blocks_count);
- }
-
- INIT_LIST_HEAD(&meta_group_info[j]->bb_prealloc_list);
-
-#ifdef DOUBLE_CHECK
- {
- struct buffer_head *bh;
- meta_group_info[j]->bb_bitmap =
- kmalloc(sb->s_blocksize, GFP_KERNEL);
- BUG_ON(meta_group_info[j]->bb_bitmap == NULL);
- bh = read_block_bitmap(sb, i);
- BUG_ON(bh == NULL);
- memcpy(meta_group_info[j]->bb_bitmap, bh->b_data,
- sb->s_blocksize);
- put_bh(bh);
- }
-#endif
-
+ if (ext4_mb_add_groupinfo(sb, i, desc) != 0)
+ goto err_freebuddy;
}
return 0;
err_freebuddy:
- while (i >= 0) {
+ while (i-- > 0)
kfree(ext4_get_group_info(sb, i));
- i--;
- }
i = num_meta_group_infos;
err_freemeta:
- while (--i >= 0)
+ while (i-- > 0)
kfree(sbi->s_group_info[i]);
iput(sbi->s_buddy_cache);
err_freesgi:
unsigned i;
unsigned offset;
unsigned max;
+ int ret;
if (!test_opt(sb, MBALLOC))
return 0;
} while (i <= sb->s_blocksize_bits + 1);
/* init file for buddy data */
- i = ext4_mb_init_backend(sb);
- if (i) {
+ ret = ext4_mb_init_backend(sb);
+ if (ret != 0) {
clear_opt(sbi->s_mount_opt, MBALLOC);
kfree(sbi->s_mb_offsets);
kfree(sbi->s_mb_maxs);
- return i;
+ return ret;
}
spin_lock_init(&sbi->s_md_lock);
ext4_lock_group(sb, md->group);
for (i = 0; i < md->num; i++) {
mb_debug(" %u", md->blocks[i]);
- err = mb_free_blocks(NULL, &e4b, md->blocks[i], 1);
- BUG_ON(err != 0);
+ mb_free_blocks(NULL, &e4b, md->blocks[i], 1);
}
mb_debug("\n");
ext4_unlock_group(sb, md->group);
-#define MB_PROC_VALUE_READ(name) \
-static int ext4_mb_read_##name(char *page, char **start, \
- off_t off, int count, int *eof, void *data) \
+#define MB_PROC_FOPS(name) \
+static int ext4_mb_##name##_proc_show(struct seq_file *m, void *v) \
{ \
- struct ext4_sb_info *sbi = data; \
- int len; \
- *eof = 1; \
- if (off != 0) \
- return 0; \
- len = sprintf(page, "%ld\n", sbi->s_mb_##name); \
- *start = page; \
- return len; \
-}
-
-#define MB_PROC_VALUE_WRITE(name) \
-static int ext4_mb_write_##name(struct file *file, \
- const char __user *buf, unsigned long cnt, void *data) \
+ struct ext4_sb_info *sbi = m->private; \
+ \
+ seq_printf(m, "%ld\n", sbi->s_mb_##name); \
+ return 0; \
+} \
+ \
+static int ext4_mb_##name##_proc_open(struct inode *inode, struct file *file)\
{ \
- struct ext4_sb_info *sbi = data; \
+ return single_open(file, ext4_mb_##name##_proc_show, PDE(inode)->data);\
+} \
+ \
+static ssize_t ext4_mb_##name##_proc_write(struct file *file, \
+ const char __user *buf, size_t cnt, loff_t *ppos) \
+{ \
+ struct ext4_sb_info *sbi = PDE(file->f_path.dentry->d_inode)->data;\
char str[32]; \
long value; \
if (cnt >= sizeof(str)) \
return -ERANGE; \
sbi->s_mb_##name = value; \
return cnt; \
-}
+} \
+ \
+static const struct file_operations ext4_mb_##name##_proc_fops = { \
+ .owner = THIS_MODULE, \
+ .open = ext4_mb_##name##_proc_open, \
+ .read = seq_read, \
+ .llseek = seq_lseek, \
+ .release = single_release, \
+ .write = ext4_mb_##name##_proc_write, \
+};
-MB_PROC_VALUE_READ(stats);
-MB_PROC_VALUE_WRITE(stats);
-MB_PROC_VALUE_READ(max_to_scan);
-MB_PROC_VALUE_WRITE(max_to_scan);
-MB_PROC_VALUE_READ(min_to_scan);
-MB_PROC_VALUE_WRITE(min_to_scan);
-MB_PROC_VALUE_READ(order2_reqs);
-MB_PROC_VALUE_WRITE(order2_reqs);
-MB_PROC_VALUE_READ(stream_request);
-MB_PROC_VALUE_WRITE(stream_request);
-MB_PROC_VALUE_READ(group_prealloc);
-MB_PROC_VALUE_WRITE(group_prealloc);
+MB_PROC_FOPS(stats);
+MB_PROC_FOPS(max_to_scan);
+MB_PROC_FOPS(min_to_scan);
+MB_PROC_FOPS(order2_reqs);
+MB_PROC_FOPS(stream_request);
+MB_PROC_FOPS(group_prealloc);
#define MB_PROC_HANDLER(name, var) \
do { \
- proc = create_proc_entry(name, mode, sbi->s_mb_proc); \
+ proc = proc_create_data(name, mode, sbi->s_mb_proc, \
+ &ext4_mb_##var##_proc_fops, sbi); \
if (proc == NULL) { \
printk(KERN_ERR "EXT4-fs: can't to create %s\n", name); \
goto err_out; \
} \
- proc->data = sbi; \
- proc->read_proc = ext4_mb_read_##var ; \
- proc->write_proc = ext4_mb_write_##var; \
} while (0)
static int ext4_mb_init_per_dev_proc(struct super_block *sb)
struct proc_dir_entry *proc;
char devname[64];
- snprintf(devname, sizeof(devname) - 1, "%s",
- bdevname(sb->s_bdev, devname));
+ if (proc_root_ext4 == NULL) {
+ sbi->s_mb_proc = NULL;
+ return -EINVAL;
+ }
+ bdevname(sb->s_bdev, devname);
sbi->s_mb_proc = proc_mkdir(devname, proc_root_ext4);
MB_PROC_HANDLER(EXT4_MB_STATS_NAME, stats);
if (sbi->s_mb_proc == NULL)
return -EINVAL;
- snprintf(devname, sizeof(devname) - 1, "%s",
- bdevname(sb->s_bdev, devname));
+ bdevname(sb->s_bdev, devname);
remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc);
remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc);
remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc);
struct ext4_sb_info *sbi;
struct super_block *sb;
ext4_fsblk_t block;
- int err;
+ int err, len;
BUG_ON(ac->ac_status != AC_STATUS_FOUND);
BUG_ON(ac->ac_b_ex.fe_len <= 0);
sbi = EXT4_SB(sb);
es = sbi->s_es;
- ext4_debug("using block group %lu(%d)\n", ac->ac_b_ex.fe_group,
- gdp->bg_free_blocks_count);
err = -EIO;
- bitmap_bh = read_block_bitmap(sb, ac->ac_b_ex.fe_group);
+ bitmap_bh = ext4_read_block_bitmap(sb, ac->ac_b_ex.fe_group);
if (!bitmap_bh)
goto out_err;
if (!gdp)
goto out_err;
+ ext4_debug("using block group %lu(%d)\n", ac->ac_b_ex.fe_group,
+ gdp->bg_free_blocks_count);
+
err = ext4_journal_get_write_access(handle, gdp_bh);
if (err)
goto out_err;
+ ac->ac_b_ex.fe_start
+ le32_to_cpu(es->s_first_data_block);
- if (block == ext4_block_bitmap(sb, gdp) ||
- block == ext4_inode_bitmap(sb, gdp) ||
- in_range(block, ext4_inode_table(sb, gdp),
- EXT4_SB(sb)->s_itb_per_group)) {
-
+ len = ac->ac_b_ex.fe_len;
+ if (in_range(ext4_block_bitmap(sb, gdp), block, len) ||
+ in_range(ext4_inode_bitmap(sb, gdp), block, len) ||
+ in_range(block, ext4_inode_table(sb, gdp),
+ EXT4_SB(sb)->s_itb_per_group) ||
+ in_range(block + len - 1, ext4_inode_table(sb, gdp),
+ EXT4_SB(sb)->s_itb_per_group)) {
ext4_error(sb, __func__,
"Allocating block in system zone - block = %llu",
block);
+ /* File system mounted not to panic on error
+ * Fix the bitmap and repeat the block allocation
+ * We leak some of the blocks here.
+ */
+ mb_set_bits(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group),
+ bitmap_bh->b_data, ac->ac_b_ex.fe_start,
+ ac->ac_b_ex.fe_len);
+ err = ext4_journal_dirty_metadata(handle, bitmap_bh);
+ if (!err)
+ err = -EAGAIN;
+ goto out_err;
}
#ifdef AGGRESSIVE_CHECK
{
spin_unlock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group));
percpu_counter_sub(&sbi->s_freeblocks_counter, ac->ac_b_ex.fe_len);
+ if (sbi->s_log_groups_per_flex) {
+ ext4_group_t flex_group = ext4_flex_group(sbi,
+ ac->ac_b_ex.fe_group);
+ spin_lock(sb_bgl_lock(sbi, flex_group));
+ sbi->s_flex_groups[flex_group].free_blocks -= ac->ac_b_ex.fe_len;
+ spin_unlock(sb_bgl_lock(sbi, flex_group));
+ }
+
err = ext4_journal_dirty_metadata(handle, bitmap_bh);
if (err)
goto out_err;
if (size < i_size_read(ac->ac_inode))
size = i_size_read(ac->ac_inode);
- /* max available blocks in a free group */
- max = EXT4_BLOCKS_PER_GROUP(ac->ac_sb) - 1 - 1 -
- EXT4_SB(ac->ac_sb)->s_itb_per_group;
+ /* max size of free chunks */
+ max = 2 << bsbits;
-#define NRL_CHECK_SIZE(req, size, max,bits) \
- (req <= (size) || max <= ((size) >> bits))
+#define NRL_CHECK_SIZE(req, size, max, chunk_size) \
+ (req <= (size) || max <= (chunk_size))
/* first, try to predict filesize */
/* XXX: should this table be tunable? */
size = 512 * 1024;
} else if (size <= 1024 * 1024) {
size = 1024 * 1024;
- } else if (NRL_CHECK_SIZE(size, 4 * 1024 * 1024, max, bsbits)) {
+ } else if (NRL_CHECK_SIZE(size, 4 * 1024 * 1024, max, 2 * 1024)) {
start_off = ((loff_t)ac->ac_o_ex.fe_logical >>
- (20 - bsbits)) << 20;
- size = 1024 * 1024;
- } else if (NRL_CHECK_SIZE(size, 8 * 1024 * 1024, max, bsbits)) {
+ (21 - bsbits)) << 21;
+ size = 2 * 1024 * 1024;
+ } else if (NRL_CHECK_SIZE(size, 8 * 1024 * 1024, max, 4 * 1024)) {
start_off = ((loff_t)ac->ac_o_ex.fe_logical >>
(22 - bsbits)) << 22;
size = 4 * 1024 * 1024;
} else if (NRL_CHECK_SIZE(ac->ac_o_ex.fe_len,
- (8<<20)>>bsbits, max, bsbits)) {
+ (8<<20)>>bsbits, max, 8 * 1024)) {
start_off = ((loff_t)ac->ac_o_ex.fe_logical >>
(23 - bsbits)) << 23;
size = 8 * 1024 * 1024;
static void ext4_mb_use_group_pa(struct ext4_allocation_context *ac,
struct ext4_prealloc_space *pa)
{
- unsigned len = ac->ac_o_ex.fe_len;
-
+ unsigned int len = ac->ac_o_ex.fe_len;
ext4_get_group_no_and_offset(ac->ac_sb, pa->pa_pstart,
&ac->ac_b_ex.fe_group,
&ac->ac_b_ex.fe_start);
if (bit >= end)
break;
next = mb_find_next_bit(bitmap_bh->b_data, end, bit);
- if (next > end)
- next = end;
start = group * EXT4_BLOCKS_PER_GROUP(sb) + bit +
le32_to_cpu(sbi->s_es->s_first_data_block);
mb_debug(" free preallocated %u/%u in group %u\n",
if (list_empty(&grp->bb_prealloc_list))
return 0;
- bitmap_bh = read_block_bitmap(sb, group);
+ bitmap_bh = ext4_read_block_bitmap(sb, group);
if (bitmap_bh == NULL) {
/* error handling here */
ext4_mb_release_desc(&e4b);
err = ext4_mb_load_buddy(sb, group, &e4b);
BUG_ON(err != 0); /* error handling here */
- bitmap_bh = read_block_bitmap(sb, group);
+ bitmap_bh = ext4_read_block_bitmap(sb, group);
if (bitmap_bh == NULL) {
/* error handling here */
ext4_mb_release_desc(&e4b);
sbi = EXT4_SB(sb);
if (!test_opt(sb, MBALLOC)) {
- block = ext4_new_blocks_old(handle, ar->inode, ar->goal,
+ block = ext4_old_new_blocks(handle, ar->inode, ar->goal,
&(ar->len), errp);
return block;
}
+ ar->len = ext4_has_free_blocks(sbi, ar->len);
+
+ if (ar->len == 0) {
+ *errp = -ENOSPC;
+ return 0;
+ }
while (ar->len && DQUOT_ALLOC_BLOCK(ar->inode, ar->len)) {
ar->flags |= EXT4_MB_HINT_NOPREALLOC;
ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS);
if (!ac) {
+ ar->len = 0;
*errp = -ENOMEM;
- return 0;
+ goto out1;
}
ext4_mb_poll_new_transaction(sb, handle);
*errp = ext4_mb_initialize_context(ac, ar);
if (*errp) {
ar->len = 0;
- goto out;
+ goto out2;
}
ac->ac_op = EXT4_MB_HISTORY_PREALLOC;
if (!ext4_mb_use_preallocated(ac)) {
-
ac->ac_op = EXT4_MB_HISTORY_ALLOC;
ext4_mb_normalize_request(ac, ar);
-
repeat:
/* allocate space in core */
ext4_mb_regular_allocator(ac);
}
if (likely(ac->ac_status == AC_STATUS_FOUND)) {
- ext4_mb_mark_diskspace_used(ac, handle);
- *errp = 0;
- block = ext4_grp_offs_to_block(sb, &ac->ac_b_ex);
- ar->len = ac->ac_b_ex.fe_len;
+ *errp = ext4_mb_mark_diskspace_used(ac, handle);
+ if (*errp == -EAGAIN) {
+ ac->ac_b_ex.fe_group = 0;
+ ac->ac_b_ex.fe_start = 0;
+ ac->ac_b_ex.fe_len = 0;
+ ac->ac_status = AC_STATUS_CONTINUE;
+ goto repeat;
+ } else if (*errp) {
+ ac->ac_b_ex.fe_len = 0;
+ ar->len = 0;
+ ext4_mb_show_ac(ac);
+ } else {
+ block = ext4_grp_offs_to_block(sb, &ac->ac_b_ex);
+ ar->len = ac->ac_b_ex.fe_len;
+ }
} else {
freed = ext4_mb_discard_preallocations(sb, ac->ac_o_ex.fe_len);
if (freed)
ext4_mb_release_context(ac);
-out:
+out2:
+ kmem_cache_free(ext4_ac_cachep, ac);
+out1:
if (ar->len < inquota)
DQUOT_FREE_BLOCK(ar->inode, inquota - ar->len);
- kmem_cache_free(ext4_ac_cachep, ac);
return block;
}
static void ext4_mb_poll_new_transaction(struct super_block *sb,
overflow = bit + count - EXT4_BLOCKS_PER_GROUP(sb);
count -= overflow;
}
- bitmap_bh = read_block_bitmap(sb, block_group);
+ bitmap_bh = ext4_read_block_bitmap(sb, block_group);
if (!bitmap_bh)
goto error_return;
gdp = ext4_get_group_desc(sb, block_group, &gd_bh);
ext4_error(sb, __func__,
"Freeing blocks in system zone - "
"Block = %lu, count = %lu", block, count);
+ /* err = 0. ext4_std_error should be a no op */
+ goto error_return;
}
BUFFER_TRACE(bitmap_bh, "getting write access");
ext4_mb_free_metadata(handle, &e4b, block_group, bit, count);
} else {
ext4_lock_group(sb, block_group);
- err = mb_free_blocks(inode, &e4b, bit, count);
+ mb_free_blocks(inode, &e4b, bit, count);
ext4_mb_return_to_preallocation(inode, &e4b, block, count);
ext4_unlock_group(sb, block_group);
- BUG_ON(err != 0);
}
spin_lock(sb_bgl_lock(sbi, block_group));
spin_unlock(sb_bgl_lock(sbi, block_group));
percpu_counter_add(&sbi->s_freeblocks_counter, count);
+ 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;
+ spin_unlock(sb_bgl_lock(sbi, flex_group));
+ }
+
ext4_mb_release_desc(&e4b);
*freed += count;